A novel reconstruction method based on RBF neural networks is proposed for restoring the degeneration Fiber Bragg Grating(FBG) strain sensor.According the correlation analysis of bridge′s FBG strain monitoring sites and none-linear approximation of RBF neural networks,the degeneration FBG strain is reconstructed by the regular and the most correlative strain sensor,which assures more accurate and accords with practice.Simulation results verify the effectiveness of the designed method and theoretical discussions.