A Stochastic Model for PLC Systems

The designers of PLC (power line carrier) systems for distributiQn networks must contend with two sources of uncertainty. First, certain topological parameters, such as transformer locations, cannot be determined accurately. Second, certain other network parameters, such as loads, equipment characteristics and circuit configurations, vary with time. A previous paper [2] has shown that signal attenuations and error rates of PLC systems are very sensitive to these uncertainties. Therefore, it is critical that they be taken into account in the design process. We develop stochastic models to aid in doing this. Specifically, the models predict the statistics of signal attenuations and error rates in arbitrary PLC systems. This is done in two stages. First, a quadratic approximation to the network´s propagation characteristics is developed. Second, this approximation and Monte Carlo sampling are used to obtain the requisite statistics.