One-bit UAV SAR Imaging Based on CDQOB Network Sub-aperture Reconstruction

Synthetic aperture radar (SAR) has been widely applied for high-resolution imaging due to its excellent characteristics. However, the large amount of data required for imaging makes it difficult to apply on resource-constrained platforms. One-bit SAR reduces data volume and alleviates the platform burden by representing echo samples as one-bit binary digital signals. Nevertheless, the high-order harmonics generated by the binary data transitions result in a degradation in imaging quality. To improve the imaging quality of one-bit SAR, this paper proposes a UAV-borne stripmap SAR imaging method based on the convolutional de-quantization onebit (CDQOB) network. The proposed method utilizes one-bit sub-aperture data to achieve motion error estimation and intelligent Range-Doppler two-dimensional spectrum reconstruction, thereby achieving high-quality stripmap SAR imaging with reduced data volume. The effectiveness of the proposed one-bit imaging method has been validated through the analysis of measured data.