基于全调节RBF神经网络的引信噪声对消研究

摘要

Fully tuned RBF neural networks can solve the uncertainties of nonlinear parameters and improve the approach ability of neural networks.To eliminate noise of fuze echo,a nonlinear adaptive filtering method based on fully tuned RBF neural networks is presented.This method can overcome the uncertainties of model and noise in signal processing.By comparing different SNR of fuze echo denoising effect,the effectiveness and superiority of fuze noise cancellation system based on fully tuned RBF neural networks is validated.%全调节RBF神经网络具有处理非线性参数的不确定性,提高神经网络的在线逼近能力,适用于噪声信号的非线性建模。为了消除引信回波中的背景噪声,提出一种基于全调节RBF神经网络的非线形自适应滤波方法,能很好克服信号处理中的模型不确定性和噪声。通过对比不同信噪比情况下引信回波的去噪效果,说明了基于全调节RBF神经网络的引信噪声对消系统的有效性和优越性。