自组织径向基神经网络在数据融合中的应用

The Application of the Radial Basis Function Network in Data Fusi on

径向基神经网络是一种基于数值计算方法插值逼近原理构造的人工神经网络 ,学习速度快且可以避免局部极小问题 .文中提出了使用一种自组织径向基人工神经网络来构造雷达功能推理机的方法 ,此网络采用非监督自组织和监督学习相结合的混合学习算法 .实验结果表明 。

Radial Basis Function Network is a kind of neural n et work, which based on interpolation-approximation theory in numeration. It has high lea rning speed and can avoid getting in local extremum .An organizing radial basis function network, which holds both non-supervised organizing function and super vised learning function, is used to construct a radar's function reasoning mecha nism. The experiment indicates that it spends much less time in training than fu zzy neural network does with the same result.