Spectrum estimation for time-series based on binary data
Résumé
This paper focuses on the spectral analysis of time series. The samples of the time series are assumed to be unknown, we only know if samples are lower or higher than a given threshold. The paper proposes an algorithm for the estimation of the (power) spectrum, this algorithm considers first the estimation of the auto-correlation function of the time series. Simulation results are given to show the effectiveness of theproposed approach.