Duty Cycle Measurement under Non Controlled Sampling: Part I

This paper discusses new algorithms for the duty cycle measurement of a digital pulse train signal that is sampled at random instants. The results show a fast convergence to the duty cycle value; in particular, a considerable gain in terms of mean number of samples nec- essary to evaluate the duty cycle with a desired accuracy is obtained, with respect to other solution based on random sampling technique already presented in literature [4]. We have developed a mathematical model valid for a generic number of collected samples and based on the hypothesis of independent and uniformly distributed samples.