An adaptive parametric time delay estimation of pulse signal via higher order statistics

Abstract

The adaptive algorithm for parametric time delay estimate of pulse signal in Gaussian noise is studied by using third order statistics. The results of computer simulations show that the approach, that used the third order cross-correlation function to solve, quite well, the problem of the time delay estimation due to the correlated noise in the case where the signal is assumed to be a nonGaussian random signal and the noise to be Gaussian (white or colored), fails to deal with the pulse signal case. A modification to the algorithm is made to estimate the time delay which is presented by the parameter of the AR model by using the adaptive algorithm. Simulations are included to show the efficiency of the modification to the pulse signal.<>