Abstract

This paper evaluates the effect of time-jitters in the equally spaced sampling wattmeters on the hypothesis of jitters uncorrelated with the input signals. The general case of two distinct time-jitters is considered, one common to the two channels and the other different for each one of them. The performance of the wattmeter has been evaluated by considering the asymptotic statistic parameters of the output. It has been shown that the different time-jitters introduce a bias and that both time-jitters contribute to the variance of the output. In any case, time-jitters introduce further bandwidth limitations which must be taken into account in the wattmeter accuracy evaluation. The theoretical results have been compared with simulated and experimental findings. Experimental results were obtained with a prototype in which both common and different time-jitters were separately added to the equally spaced sampling instants of the two input channels. In both cases, all the results were in good agreement with theoretical expectations