Efficient Behavioral Signal Path Transfer Function Evaluation

Designing an analog integrated circuit requires multiple forms of constraints. One has equalities describing circuit behavior, both DC and AC. When used in a design plan applied for optimization-based sizing, it is important that the number of analytical operations required to obtain a quantitative measure is minimal. Therefore, simplification of the analytical models allowing evaluation is desired. The goal is to simplify the model as much as possible, without loosing too much accuracy. In this paper, we will present a behavioral signal path modeling method for the AC part of a design plan. The modeling methodology transforms a device-level small-signal schematic, into a behavioral, control system equivalent. This behavioral equivalent is a model obtained by incremental modeling, which gradually simplifies the original (small-signal) circuit up to a level with sufficient accuracy over the complete design space. The model moreover allows to decompose the signal path and thus transfer function in a form allowing to evaluate the latter in a bottom up way, with a minimal amount of equation clusters or simultaneous AC constraints. Both the simplification and the way the transfer function is evaluated, make that the obtained model is computationally efficient. In addition, the behavioral signal path model offers qualitative insight into the small-signal operation of the circuit.