Detection of periodic signals in noise: an iterative procedure

A method for detection and estimation of periodic signals in the presence of noise is described. The algorithm is an iterative improvement of the autoregressive-moving average estimation of a stochastic process and gives an exact frequency resolution of sinusoidal signals additively mixed with noise in a low signal-to-noise ratio only from a small number of measurements. The iterative improvement of the spectral estimation compared with other methods is demonstrated by examples.