| 神经网络在未来超分辨雷达中的应用 |
| |
| 引用本文： | 朱兆达,佘志舜.神经网络在未来超分辨雷达中的应用[J].南京航空航天大学学报,1992(6). |
| |
| 作者姓名： | 朱兆达 佘志舜 |
| |
| 作者单位： | 南京航空学院电子工程系
(朱兆达)，南京航空学院电子工程系(佘志舜) |
| |
| 基金项目： | 国家教委博士点基金，航空科学基金 |
| |
| 摘 要： | 最新的技术发展已为研制超分辨雷达创造了条件。它能够突破普通雷达基于匹配滤波原理的分辨能力,而实现超分辨。超分辨雷达的巨大计算量通常来自求解非线性最小二乘问题时的多维搜索。本文针对这一问题,研究了雷达信号处理中的两个典型例子:(1)利用阵列处理检测个数已知而方向未知的雷达目标;(2)对波音727飞机进行超分辨距离多普勒成像。说明Hop-field神经网络通过集体运算能够求解各种困难的最优化问题,因而在未来的超分辨雷达中有广阔的应用前景。
|
| 关 键 词： | 神经网络 检测 估计 雷达阵列处理 距离多普勒成像 超分辨 |
Prospective Application of Neural Network in Superresolution Radars |
| |
| Zhu Zhaoda She Zhishun.Prospective Application of Neural Network in Superresolution Radars[J].Journal of Nanjing University of Aeronautics & Astronautics,1992(6). |
| |
| Authors: | Zhu Zhaoda She Zhishun |
| |
| Institution: | Department of Electronic Engineering |
| |
| Abstract: | Recent progresses in both theory and technology have created a chance of developing superresolution radars that may break through the radar resolution limitations imposed by the matched filter principle. The huge computational burden involved in a superresolution radar sometimes comes from a multi-dimensional search in solving a nonlinear least squares problem. Through two primary examples of detecting targets of a known number but with unknown directions using array processing and superresolution range-Doppler imaging of a commercial aircraft Boeing-727,we illustrate that the Hopfield neural network which is able to collectively compute a good solution to difficult optimization problems seems to be highly promising in prospective applications in superresolution radars. |
| |
| Keywords: | neural networks detection estimation radar array processing range-Doppler imaging superresolution |
| 本文献已被 CNKI 等数据库收录！ |
