Abstract
In many applications of signal processing, it is required to compute higher-order moments of a given random process. In this paper, we propose a new approach for estimating higher-order moments of a stationary random process given only finite data records. This approach makes use of the Discrete Orthogonal Laguerre Functions (DOLFs). Computer simulations illustrating the effectiveness of the DOLFs-based estimation methods compared to the classical estimation methods are presented.