Time delay estimation for noise-like signals embedded in non-Gaussian noise using adaptive robust DFT
Viacheslav Oliinyk, Владимир Васильевич Лукин, Igor Djurović · 2018
We consider a task of time delay estimation for two sensors that receive noise-like wideband signal embedded in non-Gaussian environment. Due to a limited interval of signal observation and noise properties, time delay estimates can be abnormal for conventional method of cross-correlation data processing. To cope with non-Gaussian noise modeled as symmetric a-stable process, robust discrete Fourier transform (DFT) can be applied at the stage of obtaining spectral estimates in both channels. We show that the use of adaptive robust estimation is expedient for the considered situation since it leads to sufficient reduction of abnormal error probability and allows method operation in conditions of limited a priori information on non-Gaussian noise characteristics.