基于深度学习的高频雷达射频干扰自动识别与抑制

高频地波雷达的探测性能极易受到射频干扰的影响，当前射频干扰抑制的研究主要是通过人工识别来逐一处理，鲜见实时自动识别与抑制射频干扰的研究。随着深度学习在雷达图像处理方面应用的展开，本文尝试将其引入高频雷达射频干扰抑制中，利用YOLO (You Only Look Once)模型来识别雷达距离多普勒谱图中的射频干扰，继而用高阶奇异值分解(Higher Order Singular Value Decomposition, HOSVD)方法对其进行抑制。仿真和实测数据处理结果表明，此YOLO-HOSVD联合算法实现了对高频雷达射频干扰的自动识别与抑制，单场数据处理时间不超过1.8 s。该方法可以应用于高频地波雷达常规海态观测。

Automatic Identification and Suppression of Radio Frequency Interference of HF Radar Based on Deep Learning

The detection performance of high frequency surface wave radar (HFSWR) is extremely susceptible to radio frequency interference (RFI). Currently, the suppression of RFI in HFSWR echoes is conducted frame by frame through manual identification, and there are few researches on automatic identification and suppression of RFI. With the development of deep learning in radar image processing, this article tries to introduce it into high frequency radar RFI suppression, and uses the YOLO (you only look once) model to identify the RFI in the radar range Doppler spectrum, and then suppressed by the higher order singular value decomposition (HOSVD) method. The results of simulation and experiment show that the YOLO-HOSVD joint algorithm can realize the automatic identification and suppression of RFI in HFSWR echoes, and the processing time of a single field data does not exceed 1.8 s. This method is applicable to the operational HFSWR to automatically identify and suppress the RFI in observing sea state.