Detecting an Unknown Abrupt Change in the Band Center of the Fast-Fluctuating Gaussian Random Process

Abstract

The generalized maximum likelihood algorithm is introduced for detecting the abrupt change in the band center of a fast-fluctuating Gaussian random process with the uniform spectral density. This algorithm has a simpler structure than the ones obtained by means of common approaches and it can be effectively implemented on the base of both modern digital signal processors and field-programmable gate arrays. By applying the multiplicative and additive local Markov approximation of the decision statistics and its increments, the analytical expressions are calculated for the false alarm and missing probabilities. And with the help of statistical simulation it is confirmed that the proposed detector is operable, while the theoretical formulas describing its quality and efficiency approximate satisfactorily the corresponding experimental data in a wide range of parameters of the observable data realization.

Keywords

• fast-fluctuating random process
• abrupt change
• maximum likelihood method
• discontinuous parameter
• unknown band center
• local markov approximation method
• false alarm probability
• missing probability
• statistical simulation