A time-domain, non-Monte Carlo method for computer simulation of electrical noise in nonlinear dynamic circuits with arbitrary excitations and arbitrary large-signal waveforms is presented. This time-domain noise simulation method is based on results from the theory of stochastic differential equations. The noise simulation method is general in the following sense. Any nonlinear dynamic circuit with any kind of excitation, which can be simulated by the transient analysis routine in a circuit simulator, can be simulated by our noise simulator in time-domain to produce the noise variances and covariances of circuit variables as a function of time, provided that noise models for the devices in the circuit are available. Noise correlations between circuit variables at different time points can also be calculated. Previous work on computer simulation of noise in electronic circuits is reviewed with comparisons to our method. Shot, thermal, and flicker noise models for integrated-circuit devices, in the context of our time-domain noise simulation method, are discussed. The implementation of this noise simulation method in a circuit simulator (SPICE) is described. Two examples of noise simulation (a CMOS inverter and a BJT active mixer) are given.

A behavioral representation of Nyquist data converters is presented. The representation captures the static behavior of a memoryless Nyquist data converter including statistical variations. The variations are classified into noise and process variations according to how these nonidealities affect the converter behavior. To describe noise effects, a joint probability density function is used. To describe behavioral effects due to process variations, a Gaussian model is used. Using the behavioral representation, a novel strategy to calculate system performance is developed. The performance specifications of a converter, including offset error, full scale gain error, integral nonlinearity, differential nonlinearity, harmonic distortion, and signal-to-noise ratio, are calculated in two steps. First, the converter model parameters are extracted from the circuit. Then, the converter performance is computed using only the model parameters since the model captures the converter behavior. Experimental results agree well with SPICE simulations and confirm the validity of the model.>

In analog system design, final verification in the presence of parasitic loading effects is crucial to guarantee functionality of the entire circuit.In this paper, we present a methodology for analog system verification in the presence of parasitic using behavioral simulation.When applied to a synthesized 10 bit D/A, our approach is accurate to 0.005 LSB compared with SPICE, while being several orders of magnitude faster.

This paper presents a strategy for testing all DC performance of Nyquist data converters including offset error, full scale gain error, integral nonlinearity, and differential nonlinearity. In contrast to previous testing strategies based on linear models that require accurate measurements of circuit performance, our strategy uses a simpler measurement to verify that a circuit performance parameter falls within certain detection thresholds in the presence of measurement noise. Using the proposed strategy, we can evaluate tradeoffs between test set size, test coverage, detection thresholds, measurement noise, chip performance, and estimated yield. Our results support the obvious that smaller measurement noise, stricter detection thresholds, and lower chip performance would require smaller test set and reduce test time. Stricter detection thresholds, on the other hand, would decrease estimated yield.

A direct noise analysis approach for mixed-mode systems is presented with experimental results compared with results from the traditional Monte Carlo approach. The direct approach computes noise effects by performing arithmetic on moments of distribution functions that characterize electronic noise. One key advantage of this approach is its ability to compute low error probabilities. From experimental results, it is shown that very low order moments, such as second order, are sufficient for a good estimate of noise effects.>

The authors present a behavioral representation for the class of Nyquist rate A/D (analog-to-digital) converters. The representation captures the nominal A/D behavior as well as all the statistical variations. The variations are classified into noise and process variations according to how these nonidealities affect the A/D behavior. To describe noise effects a joint probability density function is used. To describe behavioral effects due to process variations, use is made of a variance-covariance matrix, Sigma /sub t/, which is a generalization of the integral nonlinearity vector. Sigma /sub t/'s rank characterizes the testability of an A/D; its decomposition yields efficient strategies for A/D testing. Finally, parameter extraction results obtained from prototypes are presented.>