一种有效的雷达辐射源信号数据清洗方法

Effective data cleaning approach for radar emitter signals

为提高侦察接收机截获的辐射源数据的数据质量,提出一种针对雷达辐射源信号的数据清洗方法,能检测并剔除侦察接收机接收到的原始数据集中的噪声数据,提高数据质量,降低低质量数据或噪声数据对电子侦察系统后续处理流程的影响。首先对辐射源信号进行预处理,获取信号包络;之后提取辐射源信号包络的相像系数作为特征向量;最后用基于密度的含噪数据空间聚类算法(DBSCAN)对特征数据进行聚类,区分噪声数据和有效脉冲数据,将噪声数据剔除,从而提高数据质量。在仿真数据和实测数据上的实验证明,提出的数据清洗方法基本达到了人工清洗的水平,能有效提高数据质量和清洗效率。

In order to improve the data quality of radar emitter signals intercepted by reconnaissance receiver,a data cleaning method for radar emitter signals is proposed,which can detect and eliminate the noise data(in the original data set)received by the reconnaissance receiver,improve the data quality and reduce the influence of low⁃quality data or noise data on the follow⁃up processing flow of electronic reconnaissance system.The radar emitter signals are preprocessed to get the signal envelope first,and then the resemble coefficients of the emitter signal envelope are extracted as the feature vector.Finally,the feature data is clustered with DBSCAN(density⁃based spatial clustering of applications with noise).In this way,the noise data is distinguished from the effective pulse data and removed from the data set,so that the data quality is improved.The experiments with the simulation data and the measured data show that the data cleaning method proposed has basically reached the level of manual cleaning,and can improve data quality and cleaning efficiency effectively.