基于图像分割的高频雷达射频干扰提取算法

高频雷达必需射频干扰抑制,而抑制效果常受制于干扰样本质量。针对雷达正常工作如何提取干扰样本的问题,首次提出基于距离-多普勒(range-Doppler, RD)图图像分割的射频干扰提取方法。通过将常规处理的RD图转换为灰度图像,然后设计去海杂波、边缘检测、分段和整段直线检测算法,从而确定干扰边缘线并提取出干扰区域。由此恢复的干扰数据,不受海杂波影响,能够准确估计干扰特性和提高抑制效果。实践表明,该算法准确度高,能提取被海杂波覆盖达75%的干扰线。为验证其应用,采用该算法辅助改进滤波器设计,能够取得良好的干扰抑制效果。

Extraction of radio frequency interference based on image segmentation for high frequency radar

High frequency radars need to suppress the radio frequency interference (RFI), where suppression performances are constrained by RFI samples. To obtain the RFI samples from the received data or the range-Doppler (RD) map, a novel approach is proposed to extract the RFI regions in the RD map based on image segmentation methods. By transforming the RD map into a gray image, algorithms of sea clutter mitigation, edge detection, segments and lines detection are developed, to extract the RFI edges and regions. Thus, the recovered RFI data can be used to estimate the RFI properties accurately and improve the suppression performances. Practical experiences show that the proposed approach can detect the RFI line successfully even when 75% of it is covered by sea clutter. To demonstrate its application, the proposed algorithms is also applied in receive filter design which can suppress the RFI effectively.