Application of array processing techniques to multibaseline InSAR for layover solution

Synthetic aperture radar interferometry (InSAR) is a modern technique to derive digital height maps of the land surface from SAR images. In recent years there has been great interest in exploiting the advanced multibaseline operation for solving layover effects that can degrade SAR and InSAR imagery. We consider detailed modelling of this problem including speckle noise for extended targets, application of several nonparametric and parametric spectral estimation methods to multibaseline layover solution, and performance analysis. The problem of layover solution is formulated as the estimation of a multi-component signal composed by multiple cisoids corrupted by complex multiplicative Gaussian noise and additive white Gaussian thermal noise. Beamforming, Capon's, High-Order Yule-Walker, MUSIC, Min-Norm, and ESPRIT spatial spectral estimators are applied to this interferometric problem, and an extensive simulated performance comparison is carried out in terms of statistical accuracy and resolution.