单比特技术在合成孔径雷达成像上具有良好的抗噪性,大多数单比特重建算法使用L1或L2范数作为稀疏正则化。文中提出了能获得更好稀疏解的L1/2范数,使用交替方向乘子法对回波信号进行重构,通过仿真实验和实际回波图验证了其性能。相较于固定阈值,变阈值能保留目标反射系数的一部分幅值信息,因此文中进一步使用了自适应阈值来最大程度地接近回波数据。仿真结果显示:与时变阈值算法相比,恢复出的图像与原图像的误差更小,并且在L1/2范数下具有更加明显的效果,在低信噪比时,自适应阈值相较于时变阈值也展现出了更好的性能。

1-bit technique has good noise immunity on synthetic aperture radar imaging. Most 1-bit reconstruction algorithms use L1 or L2 norm as sparse regularization. In this paper, a L1/2-norm that can obtain better sparse solutions is proposed, an alternating direction multiplier method is used to reconstruct the echo signal, and its performance is verified by simulation experiments and actual echo diagrams. Compared with the fixed threshold, the variable threshold can retain partial amplitude information of the target reflection coefficient. Thus an adaptive threshold is further used to approximate the echo data to the greatest extent. The simulation results show that, compared with the time-varying threshold algorithm, the error between the restored image and the original image is smaller, and a more obvious effect under the L1/2 norm is gotton. When the signal-to-noise ratio is low, the adaptive threshold also performs better than the time-varying threshold.

TN957.51