In this chapter, the concept of noise is introduced both as a stochastic process—defined as a sequence of random variables—and as a perturbative element within a dynamical system. A classification of noise is then provided based on its frequency content, commonly referred to by its “color”, and a distinction is made between output noise and dynamical noise. The latter, being embedded within the system’s evolution, fundamentally alters its dynamics and can be interpreted as information intrinsic to the system that cannot be inferred solely from its past behavior. The chapter focuses primarily on dynamical noise, presenting a survey of existing methods for its estimation within dynamical systems. Main approaches including nonlinear time series analysis, stochastic differential equations, probabilistic-Bayesian frameworks, and state-space modeling, including the well-known Kalman filter, are briefly reviewed.