Near-Field Radar Imaging Motion Compensation Solution Based On STFT and PGA

With the continuous advancement of Terahertz radar technology, high-resolution, high-definition and real-time Terahertz near-field imaging has emerged as a new research objective. However, due to the radar's high-speed motion and the accompanying mechanical vibrations, significant motion errors are induced, which in turn substantially deteriorate the imaging quality. Conventional studies often simplify vibration to a sinusoidal model, but this assumption may not fully reflect real-world conditions. This paper proposes a more realistic perturbation model and a motion compensation scheme based on STFT and PGA, which can better address nonlinear problems. Through actual experiments and ablation studies, this paper demonstrates that this method has the advantages of fewer iterations and better imaging quality. Under the same image contrast, the algorithm proposed in this paper only requires 3 iterations. The PSNR of the imaging results of the algorithm proposed in this paper increases by approximately 3 dB.