| 基于径向基神经网络的波达方向估计算法 |
| |
| 引用本文： | 何宏,李涛,张志宏,杨桐,何林. 基于径向基神经网络的波达方向估计算法[J]. 徐州工程学院学报, 2012, 0(3): 6-10 |
| |
| 作者姓名： | 何宏 李涛 张志宏 杨桐 何林 |
| |
| 作者单位： | [1]天津理工大学天津市复杂系统控制理论及应用重点实验室,天津300384 [2]天津广播电视发展有限公司,天津300072 [3]天津移动通信有限责任公司,天津300052 |
| |
| 基金项目： | 天津市科技创新专项资金项目（10FDZDGX00400） |
| |
| 摘 要： | 针对传统的波达方向（DOA）估计算法在实际应用中普遍存在计算量较大,无法实时地跟踪期望信号且无法处理信号源数大于天线阵元数的问题,提出了一种在智能天线中基于径向基神经网络的波达方向估计算法.该算法利用神经网络进行多信号源跟踪（MUST）来完成信号源侦测和DOA估计.通过建立神经网络DOA估计算法模型,并对所建立的神经网络进行训练.通过仿真将该算法与传统的DOA估计算法进行比对的结果表明,基于径向基神经网络的波达方向估计算法能够快速准确的检测到信号源,响应时间明显快于传统的算法.
|
| 关 键 词： | 智能天线 概率神经网络 广义回归网络 波达方向估计 径向基网络 |
Direction of Arrival Estimation Algorithm Based on Radial Basis Function Neural Network |
| |
| HE Hong,LI Tao,ZHANG Zhi-hong,YANG Tong,HE Lin. Direction of Arrival Estimation Algorithm Based on Radial Basis Function Neural Network[J]. Journal of Xuzhou Istitute of Technology, 2012, 0(3): 6-10 |
| |
| Authors: | HE Hong LI Tao ZHANG Zhi-hong YANG Tong HE Lin |
| |
| Affiliation: | 1. Tianjin Key Laboratory for Control Theory and Application in Complicated Systems, Tianjin University of Technology, Tianjin 300384, China; 2. Tianjin Radio and Television Group Development Co. ,Ltd, Tianjin 300072, China;3. Tianjin Mobile Communications Co. , Ltd, Tianjin 300052, China) |
| |
| Abstract: | Against the problem of the large amount of calculation and inability to track targeted sources real-time and to locate sources that are greater than the number of array elements number with traditional DOA estimation algorithm, a new direction of arrival estimation algorithm based on radial basis function neural network （RBFNN） in smart antenna is proposed in this paper. The proposed neural multiple-source tracking （N-MUST） algorithm is based on probabilistic neural network and general regression neural net- work to perform both detection and direction of arrival estimation. The model of neural network in direc- tion of arrival estimation is established and trained in this paper. Simulation results, which compared the traditional algorithm and the new one, indicate that the direction of arrival estimation algorithm based on RBFNN implement multiple-source tracking exactly and fast. |
| |
| Keywords: | smart antenna probabilistic neural network general regression neural network DOA es- timation radial basis function neural network |
| 本文献已被 维普 等数据库收录！ |
|
