Cumulant-based Adaptive Time Delay Estimation

A new adaptive scheme for time delay estimation is introduced for signal environments where the signal is non-Gaussian and the additive noise sources are spatially correlated Gaussian with unknown power spectrum characteristics. The new scheme is based on parametric modeling between two sensor measurements and employs higher-order statistics (third- or fourth-order) of the data. An analytic asymptotic performance evaluation of the errors of the model parameters obtained by the new method is provided. It is demonstrated by means of extensive simulations that the new adaptive scheme works well for both stationary and nonstationaxy cases. As expected, it outperforms the cross- correlation-based gradient method for time delay adaptation in spatially correlated Gaussian noises. The new scheme is also compared with the overdetermined recursive instrumental variable method and shown to exhibit substantially less computational complexity.