Intra-Pulse Feature Extraction of Radar Emitter Signals Based on Complex Network

The electromagnetic signal environment of the modern battlefield is complex. The radar signal recognition technology based on the traditional five parameters cannot meet the actual needs. For this reason, it is urgent to explore new identification methods to improve the technical level of existing radar countermeasure equipment in electronic warfare. Aiming at the research challenge of low-accuracy intra-pulse waveform recognition in a dense modern electronic warfare environment, In this paper, we propose a new method to extract the characteristics of radar emission signals. This method first reconstructs the radar transmitter signal into a complex network, and then extracts the characteristics of the complex network. In our algorithm, the features are extracted from network topology statistics with each vector point of the reconstructed phase space represented by a single node and edge determined by the phase space distance. Through analyzing the global properties of network nodes, we found that the network constructed by the method described in this article inherited the dynamic characteristics of all aspects of the intentional intra-pulse modulation type of radar signals in time domain. Furthermore, we investigate the stability and sensibility of feature parameters with the variation range from 5dB to 20dB.

Computer simulation demonstrates how the topological indices of the network can be used to distinguish different modulation types of radar emitter signals. At the same time, the experimental results show that the extracted feature vectors have good ability of noise-resistance and good clustering quality in low SNR environment when the radar signals are corrupted by measurement noise.