Time delay estimation in unknown spatially uncorrelated Gaussian noises using higher-order statistics

Abstract

Bispectrum methods have been proposed for non-Gaussian signal time delay estimation (TDE) problem. When the signal is non-Gaussian and additive noises are spatially uncorrelated Gaussian, the bispectrum methods are outperformed by generalized cross-correlation (GCC) methods. This problem is addressed in this paper and new methods are proposed to improve the TDE performance. The new methods exploit the higher-order statistics characteristics of the signals and formulate weighting functions to improve the time delay estimation. Computer simulation results show that the new methods outperform both the GCC and the bispectrum methods.