基于时频域分析的车载毫米波雷达干扰抑制方法

Interference Mitigation for Automotive millimeter wave Radars Based on Time-Frequency Domain Analysis

文中针对车载调频连续波雷达之间相互干扰导致虚警和漏警的问题,提出一种在时频域基于改进经验模式分解和自回归模型相结合的干扰抑制方法。该方法首先使用经验模式分解区分出拍频信号中干扰分量主导的低阶本征模态函数,将其转换到短时傅里叶变换域后通过全局阈值方法进行干扰分量定位;其次,在时频域根据定位信息将拍频信号包含干扰的数据置零;最后,使用自回归模型对拍频信号中缺失的有用信号进行估计并插值。通过仿真和实测结果显示,该方法在精确地去除干扰分量的同时可以减少有用信号的功率损失,干扰抑制后的信号与参考信号的相关系数达到0.9697。与现有干扰抑制技术相比文中方法也体现出更优的干扰抑制性能。

The paper addresses the issue of mutual interference between frequency modulated continuous wave radar systems leading to false alarms and missed detections.It proposes an interference mitigation method based on improved empirical mode decomposition and autoregressive models in the time-frequency domain.The method first employs empirical mode decomposition to distinguish low-order intrinsic mode functions dominated by interference components in the beat frequency signal.These components are then transformed into the short-time Fourier transform domain,and interference components are located using a threshold method.Subsequently,in the time-frequency domain,the data containing interference in the beat frequency signal is set to zero based on the localization information.After eliminating the interference components,an autoregressive model is used to estimate and interpolate the missing useful signals in the beat frequency signal.Simulation and experimental results demonstrate that this method can accurately remove interference components while reducing the power loss of useful signals.The correlation coeffi-cient between the signal after interference suppression using this method and the reference signal is 0.9697.Compared to existing interference mitigation techniques,the proposed method also exhibits supe-rior interference mitigation performance.