Integrated Circuit Modeling for Noise Susceptibility Prediction in Communication Networks

This paper addresses an integrated circuit (IC) modeling procedure for mixed-signal immunity simulations of communication networks. The procedure is based on a gray-box approach, modeling (IC) ports with a physical circuit and the internal links with a behavioral block. The parameters are estimated from time and frequency domain measurements, allowing accurate and efficient reproduction of nonlinear device switching behaviors. The effectiveness of the modeling process is verified by applying the proposed technique to a controller area network (CAN) transceiver, involved in a direct power injection (DPI) immunity test on a data communication link. The obtained model is successfully implemented in a VHDL-analog mixed-signal (AMS) solver to predict both the functional signals and the RF noise immunity at component level.