Multi-window spectral estimation with optical processing for ISAR imaging

We propose a new method for adaptive nonstationary spectral estimation and harmonic analysis for stepped-frequency inverse synthetic aperture radar (ISAR) imaging. The essential idea is to construct a set of orthogonal windows that have little spectral leakage and use each of these windows to create an independent spectral estimate. A data dependent and frequency dependent weighted average of these individual estimates provides the final estimate. Since this procedure has a strong theoretical basis and has enjoyed great success in geophysical applications, it is almost guaranteed to significantly improve ISAR imaging. In addition. multi-window spectral estimation seems suited to optical processing. in Phase 1, windows matched to a set of simulated radar returns will be computed. Computer programs for spectral estimation will be created. In Phase I Option, multiwindow estimation will be compared with single-window estimation. and if multiple windows are significantly better for imaging, we will collaborate on integrating the multi-window algorithm into digitally and optically based ISAR imaging systems. Although the algorithms developed and tested will be used to analyze radar returns. they may be used for the detection and analysis of any nonstationary signal in additive broadband nongaussian noise,