Denoising Method of Pulsar Photon Signal Based on Recurrent Neural Network

In order to obtain the pulse contour with high signal to noise ratio in short observation time, a denoising method of pulsar photon signal based on recurrent neural network is proposed, and the actual observation data of Rossi satellite in the United States are used to analyze and verify the method. Using signal to noise ratio and mean square variance as the evaluation index, the results of measured data show that recurrent neural network has a significant effect on reducing the noise contained in the observational contour of pulsars. Compared with the traditional method of periodic folding, the proposed method can greatly reduce the observation time needed to obtain the contour of the same signal to noise ratio, and is more conducive to realtime application.