Fluctuation analysis for a class of nonlinear systems with fast periodic sampling and small state-dependent white noise (Q6042668)

The dynamics of a nonlinear differential equation with fast periodic sampling under the influence of small white-noise perturbations is analyzed in this paper. For the resulting stochastic process indexed by two small parameters, the authors compute the zeroth and first order terms for a perturbation expansion in terms of the small parameters, along with rigorous estimates on the ensuing error. The zeroth order term, describing the mean dynamics, is given by a limiting ordinary differential equation (ODE), while the first order term, capturing fluctuations about the mean, is given by a linear non-homogeneous stochastic differential equation (SDE). This SDE is found to vary depending on the relative rates at which the two small parameters approach zero. An effective drift term in the limiting SDE for the fluctuations has been identified, capturing the combined effect of small noise and fast sampling. Finally, the theoretical results have been illustrated numerically in the context of a control problem. Several avenues for future exploration are suggested by the problems studied here. Explicit solution representations for the state process and the possibility of linking the analysis to some limit theorems for products of random matrices were utilized in these calculations. It would be of interest to investigate whether these results can be extended to encompass the full continuous-time state process for both linear and nonlinear systems.