A Spectral Clustering Sorting Method For Radar Signals Based On SOM Anchor Extraction And Graph Fusion

As a key technology of reconnaissance system, radar signal sorting (RSS) aims to separate the pulse sequence emitted by different radar from received pulse stream. However, with the emergence of multi-function radars (MFRs), the complex distribution of pulses presents challenges for radar signal sorting in the battlefield electromagnetic environments. A spectral clustering sorting method for radar signal (WFG-SC) is proposed to tackle the problem mentioned above. Our method utilizes Self-Organizing Map (SOM) and K-nearest neighbors (KNN) to construct an adaptive topological graph of the pulse stream. Then, the spectral clustering algorithm is employed for radar signal sorting. Simulation results on MFR pulse dataset show that our method improves the performance of spectral clustering and outperforms some existing sorting methods.