Efficient ISAR Range Envelope Alignment via Phase Difference
Abstract
Inverse synthetic aperture radar (ISAR) serves as an effective all-day, all-weather technique for imaging noncooperative targets. Following range compression of the ISAR signal, subsequent processing typically involves range envelope alignment, initial phase compensation, rotational motion compensation, and azimuth imaging. As the first critical step, range envelope alignment aims to maintain the echo from each individual scatterer within its respective range bin throughout the imaging interval, with its precision directly impacting subsequent phase and motion compensation. Conventional range alignment methods rely exclusively on amplitude-based techniques—such as peak detection, centroid tracking, correlation, or entropy and contrast metrics. In contrast, this article proposes a novel, highly efficient envelope alignment method utilizing phase differences. The approach first divides the range spectrum at each azimuth sample into blocks, generating two signals from two spectral segments per azimuth sample. Subsequently, for four adjacent signals across two consecutive samples, three conjugate multiplications are performed, and the relative offset is estimated via maximum likelihood estimation (MLE). The estimated offset is then applied to align the envelopes of adjacent samples. Finally, the aligned signals are phase-adjusted and coherently integrated before proceeding to align the next sample iteratively. Compared with conventional methods, the proposed noninterpolation approach achieves comparable alignment accuracy while reducing computational complexity and random access memory footprint. Furthermore, its rapid, pulse-to-pulse processing architecture makes it highly suitable for real-time applications in high-pulse repetition frequency (PRF) radar operational modes. Validation using both simulated and measured data confirms the effectiveness of the proposed technique.