Aiming at characteristics of noise in tank cabin

design a tank helmet model which is based on improved RBF neural network active denoising.The principle is on the basis of using RBF neural network to dispose noise

add spectrum analysis theory to construct delay reverse noise

after stacking with secondary sound source signal produced by the filtering algorithm of RX-RBF

interference cancellation is carried out on noise source.As the space of tank helmet is so small that the scheme ignores space inconsistent problem caused by original noise

but it effectively solve time inconsistent problem caused by anti-noise source and original noise.Simulation by Matlab verify that the proposed scheme not only reflects characteristics of fast training speed and good local approximation effect of RBF neural network

but also effectively improve denoising effect.