Multitaper ISAR Noise Suppression

This article presents a robust technique for inverse synthetic aperture radar (ISAR) image enhancement using bivariate multitaper spectrum estimation with simultaneous concentration. The proposed method aims to average multiple focused ISAR images obtained using orthonormal windows. As a result, an image with an improved concentration and reduced noise is obtained. Compared to state-of-the-art methods, the presented approach efficiently suppresses stationary and non-stationary noise in ISAR images, leading to contrast and entropy enhancement. Extensive simulations are shown to illustrate the superiority of the proposed technique quantitatively, confirming its utility in practical systems. Ultimately, the real-life high-resolution ISAR image processing result is shown, demonstrating the method’s performance in active frequency-modulated continuous wave (FMCW) radars.