1: \begin{abstract}

2: Data observed at high sampling frequency are typically assumed to be an additive composite of a relatively slow-varying continuous-time component,  a latent stochastic process or a  smooth random function,  and measurement error.

3: Supposing that the latent component is an It\^{o} diffusion process,

4: we propose to estimate the measurement error density function by applying a deconvolution technique with appropriate localization.

5: Our estimator, which does not require equally-spaced observed times, is consistent and minimax rate optimal.  We also investigate estimators of the moments of the error distribution and their properties,

6: propose a frequency domain estimator for the integrated volatility of the underlying stochastic process, and show that it achieves the optimal convergence rate.

7: Simulations and a real data analysis validate our analysis.

8: \end{abstract}