Some considerations about the integration of r-ACF

The integration of the autocorrelation function (r-ACF) over finite temporal windows presents well known signal analysis problems. This work shows how it is possible to perform the computation taking advantage from acknowledged methods and at the same time avoiding common problems. The aim of this work is to suggest how to obtain the most accurate results, optimizing the computation algorithms. In particular, this study investigates how much different r-ACF integration methods affect the computed values of τe. The main results of previous r-ACF studies are discussed, then different integration solutions are applied to two reference sample tracks, analyzing the corresponding significant factors. Finally, different optimization techniques of the integration parameters are presented and compared, with particular attention to their computational load.