Statistical DC node voltage characteristics by Monte-Carlo techniques

The calculation of the standard deviations and mean values of the node voltages of a given network by ECAP are based upon the assumptions that a node voltage e[subscript k] is a linear function of the circuit variables P[subscript j] such that e[subscript k]=[sum][subscript j]K[subscript j]P[subscript j], and that all variable parameters are distributed. Regarding the first assumption, e[subscript k] normally realistically cannot be expected to be linear with respect to all its variable parameters. The second assumption does not hold in some cases where, for example, the variable parameters may be assummed to have uniform distribution. In either case, ECAP analysis will generally not yield the actual mean and standard deviation of node voltages. Therefore it was desirable to modify the ECAP program so that the user may get a more accurate estimate of the mean values and standard deviations. To fulfill this requirement, a computer program using the Monte-Carlo method has been developed. This method generates normally distributed random variable parameters of the network and then calculates the node voltages. From these, in turn, the mean values and standard deviation of node voltages are calculated.