摘要
在空间分割的基础上,提出了采用四个独立的二次规划神经网络neuralnetworks(NN)实现方向估计的方法。通过将空间角度分割成许多细小的等分,从这些不同的角度估计出沿该角度入射信号的幅度,从而判断出目标信号的个数和方向,利用该方法设计的网络具有高速、实时的特点,计算机模拟结果表明该方法很有效。
In this paper four neural networks for solving quadratic optimization problems are used for direction finding based on spatial partition.The amplitude of arrival signal in each one of variant directions is estimated.According to the to the estimated amplitudes of arrival signals in variant directions,the mumber and directions of the true arrival signals can be found .The network given in the paper has the satisfactory performance.
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
1995年第4期355-359,共5页
Journal of University of Electronic Science and Technology of China
基金
安徽省教委基金
关键词
神经网络
方向估计
空间分割
neural network
bearing estimation
spatrial partition