<p>Noise can be modeled as a sequence of random variables defined on a probability space that may add to a given dynamical system T that is a map on a phase space. In the nontrivial case of dynamical noise {ε<sub>n</sub>}<sub>n</sub>, where ε<sub>n</sub> ∈ N(0, σ<sup>2</sup>) and the system output is x<sub>n</sub> = T(x<sub>n−1</sub>, x<sub>n−2</sub>, ..., x<sub>0</sub>)+ε<sub>n</sub>, without any specific knowledge/assumption on T, the quantitative estimation of noise power σ<sup>2</sup> is a challenge. Here, we introduce a formal method, based on nonlinear entropy profile, to estimate the dynamical noise power σ<sup>2</sup> without knowledge on the specific T function. Time-series generated from Logistic maps and Pomeau-Manneville systems under different conditions are used to test the correctness of the method. Results demonstrate that the proposed estimation algorithm properly discerns different noise levels with no a priori information. </p>
<p>Numerical simluations, including sereis generation, are performed with MATLAB.</p>