A Recovery Method for Abnormal Radar Parameters Based on Improved VAE Model

To address the problem of abnormal radar parameters caused by electromagnetic interferences, a recovery method for abnormal radar parameters based on improved Variational Autoencoder model with self-attention (SA-VAE) is proposed. By incorporating a self-attention mechanism into the encoder of the VAE, the model can capture the complex internal dependencies of radar parameters and enhance the feature extraction capabilities of the model. Simulation results indicate that the recovery deviation of abnormal radar parameters using the proposed method is consistently within 5%, which demonstrates that the SA-VAE model has higher reconstruction accuracy compared to the original VAE model.