The linear time-varying frequency errors (LTFE) caused by the mismatch of transmitter and receiver oscillators can defocus the imaging result of distributed inverse synthetic aperture radar (ISAR) seriously. The LTFE calibration method based on the entropy minimization principle is sensitive to signal-to-noise ratio (SNR), and its performance is degraded significantly under low SNR conditions. In addition, this method uses enumeration algorithm to solve the optimization problem, which has a heavy computation burden. Therefore, a robust calibration method based on the contrast maximization principle is proposed. Compared with image entropy, image contrast has better anti-noise ability because it has better sensitivity property, namely, the change of image contrast is sharper than the change of image entropy. In the proposed method, the estimation of frequency error coefficient is modelled as an unconstrained optimization problem with image contrast as cost function, and the particle swarm optimization (PSO) algorithm is used to search the global optimal solution. Then, the LTFE can be calibrated by the estimated frequency error coefficient. The proposed method has better robustness, which can work well under low SNR conditions. Besides, it has higher computational efficiency. Simulations are carried out to verify the effectiveness and robustness of the proposed method.

中文翻译：

---

基于对比度最大化原理的分布式ISAR时变频率误差的鲁棒校准方法

由发射器和接收器振荡器的不匹配引起的线性时变频率误差（LTFE）会严重散焦分布式逆合成孔径雷达（ISAR）的成像结果。基于熵最小化原理的LTFE校准方法对信噪比（SNR）敏感，并且在低SNR条件下其性能会大大降低。另外，该方法采用枚举算法解决了优化问题，计算量大。因此，提出了一种基于对比度最大化原理的鲁棒校准方法。与图像熵相比，图像对比度具有更好的灵敏度特性，具有更好的抗噪能力，即图像对比度的变化比图像熵的变化更锐利。在建议的方法中，将频率误差系数的估计建模为以图像对比度为代价函数的无约束优化问题，并采用粒子群算法（PSO）搜索全局最优解。然后，可以通过估计的频率误差系数来校准LTFE。该方法具有较好的鲁棒性，可以在低信噪比条件下很好地工作。此外，它具有较高的计算效率。通过仿真验证了所提方法的有效性和鲁棒性。该方法具有较好的鲁棒性，可以在低信噪比条件下很好地工作。此外，它具有较高的计算效率。通过仿真验证了所提方法的有效性和鲁棒性。该方法具有较好的鲁棒性，可以在低信噪比条件下很好地工作。此外，它具有较高的计算效率。通过仿真验证了该方法的有效性和鲁棒性。