方位多通道技术是合成孔径雷达(SAR)实现高分宽测的手段之一。在多通道系统中通道失配是不可避免的,这会导致SAR图像模糊。已有的通道失配校正方法大多依赖于系统参数以及场景内容。参数的不确定性将会大大降低校正算法的稳定性。该文提出了一种改进的通道失配校正方法,根据失配产生的原因,将通道失配分为距离增益误差、脉冲采样时钟误差和传输相位误差3项。前两项误差通过交替估计进行补偿,而传输相位误差则通过代价函数给予估计。该方法对成像场景的依赖较小,基于机载多通道验证平台实测数据的实验验证了该方法的有效性。 Azimuth multi-channel technology is one of the means by which synthetic aperture radar (SAR) can realize high-range-wide measurement. Channel mismatch is inevitable in multi-channel systems, which can result in ambiguous SAR images. Existing channel mismatch correction methods rely mostly on system parameters and scene content. The uncertainty of the parameters will greatly reduce the stability of the calibration algorithm. In this paper, an improved channel mismatch correction method is proposed. According to the causes of mismatch, the channel mismatch is divided into three categories: distance gain error, pulse sampling clock error and transmission phase error. The first two errors are compensated by alternating estimates, while the transmission phase error is estimated by the cost function. The method has less reliance on the imaging scene. The experimental results show that the proposed method is validated based on the experimental data of airborne multi-channel verification platform.