Stabilizing Mean Values and Minimizing Variances of Electronic Systems Performance Criteria

Mathematical statistics are used to design electronic systems in which the effects of the initial variations of the system parameters and the drift of these parameters on the system performance criteria are a minimum. This is accomplished by minimizing the initial variances of the system performance criteria with respect to the initial mean values of the system parameters while constraining the means of these parameters to their required values. In addition, the drift rates of the system parameters are selected in such a manner as to prevent the drift of the mean values of the system performance criteria while reducing the tendencies of the variances of these criteria to increase with time. These techniques were successfully applied to a number of electronic systems, including systems with both active and passive elements, thus illustrating the usefulness of mathematical statistics in the design of mass produced electronic systems.