Abstract: Real-time performance and cost constraints are critical factors limiting the large-scale deployment of foreign object debris (FOD) detection radars on airport runways. To address this challenge, this study presents a zero-intermediate-frequency (zero-IF) W-band arc-scanning synthetic aperture radar (SAR) system that combines high real-time processing capability with low power consumption, enabling rapid acquisition of high-resolution SAR images in complex airport environments. To meet the stringent requirements of real-time operation and low power consumption, this paper investigates SAR imaging algorithms. Conventional back-projection (BPA) and range-Doppler (RD) algorithms suffer from high computational complexity and excessive memory usage. To overcome these limitations, we propose an improved RD method: A joint signal and motion geometry model is established to analytically derive the coupling between range cell migration (RCM) and Doppler frequency, enabling precise phase compensation for RCM correction. Leveraging the narrow spectral support of the azimuth response function, an energy-approximation-based support region estimation method is introduced, followed by Chirp-Z transform (CZT) for efficient azimuth integration. Experimental results demonstrate that, compared to the conventional Range-Doppler (RD) imaging method, the proposed algorithm reduces memory usage by 90% and computational complexity by 67%, while achieving lower background noise in the reconstructed images. This significantly relaxes hardware platform requirements.