Radiation emitter signal recognition based on VMD and IALO-SVM

With the continuous development of modern electronic warfare technology, the enemy radar signals captured in complex environments usually have very few useful signals. In order to improve the recognition accuracy and anti-noise performance of radar emitter signal(RES), In this paper, an RES identification method based on variational mode decomposition (VMD) and improved ant lion optimization (IALO) is proposed to optimize support vector machine(SVM) parameters. Firstly, the VMD algorithm is used to decompose the radar signal into six intrinsic mode functions (IMF). Correlation coefficients were used to distinguish correlative modes and uncorrelative modes, the dominant component of noise was denoised locally, and the information dimension(DI), fractal box dimension(DF) and weighted percolation entropy(WPE) were extracted from the reconstructed signals for feature fusion to form three-dimensional feature vectors. Finally, the SVM with optimized IALO parameters is used to identify RES. The simulation results show that the method can still get a high recognition rate under the condition of low signal-to-noise ratio(SNR), when the SNR is not less than 0dB, the recognition rate can reaches 100%, and the method has a strong anti-noise performance.