Research on Target Identification Technique of Anti-ship Missile Terminal Guidance Based on GA-BP

Aiming at the target recognition problem of anti-ship missile terminal guidance radar, the statistical mean, maximum value, minimum value, range, standard deviation, variation coefficient, smoothness of RCS frequency sequence are proposed as the feature vector, which is classified by GA-BP neural network. The simulation results show that the GA-BP model greatly reduces the possibility that the BP network would be trapped in the local minimum. At the same time, compared with the single RCS size feature, the multi-dimensional RCS frequency statistical feature can significantly improve the anti-missile target recognition capability.