中文摘要：

一条新途径被介绍消除利用合成的孔雷达(SAR ) 图象设置了的仅仅单个看建筑群(SLC ) 的导致空气的阶段错误。这个方法利用了在图象象素和估计包含的阶段错误部件的空间不变性特征阶段错误用加权最少平方(WLS ) 过滤器。实际上，这种方法能作为通常在在空中的 SAR 2-D 成像被使用补偿飞机介绍的阶段错误的 autofocus 算法被分类非理想的运动。处理的真实数据，与 Honda 中心和阿纳翰测试地点的安吉尔体育场相关并且从 2004 年 6 月在时期期间由 Envisat-ASAR 获得到 2007 年 10 月，被执行评估这个 WLS 评价算法。试验性的结果证明阶段错误从 WLS 估计过滤器是很精确的，沿着 NSR 尺寸的集中的质量经由阶段修正显著地被改进，它验证这个新方法的有实行可能。

英文摘要：

A new approach was presented to eliminate the atmosphere-induced phase error utilizing only the single look complex（SLC） synthetic aperture radar（SAR） image set. This method exploited the space-invariance characteristic of phase error components contained in image pixels and estimates the phase error using the weighted least-squares（WLS） filter. Actually, this sort of method can be classified as autofocus algorithm which was generally applied in airborne SAR 2-D imaging to compensate the phase error introduced by airplane＇s nonideal motion. Real data processing, which is relevant to Honda center and Angel stadium of Anaheim test-sites and acquired by Envisat-ASAR during the period from June 2004 to October 2007, was carried out to evaluate this WLS estimation algorithm. Experimental results show that the phase error estimated from WLS filter is very accurate and the focusing quality along NSR dimension is improved prominently via phase correction, which verifies the practicability of this new method.