引用本文: | 郭小康, 简涛, 董云龙. 基于LVQ网络优化的雷达目标识别算法[J]. 雷达科学与技术, 2019, 17(1): 33-36.[点击复制] |
GUO Xiaokang, JIAN Tao, DONG Yunlong. Radar Target Recognition Algorithm Based on Optimized LVQ Network[J]. Radar Science and Technology, 2019, 17(1): 33-36.[点击复制] |
|
摘要: |
目标一维距离像在雷达目标识别领域中具有很高的研究价值,神经网络有很强的自适应能力,被广泛应用于目标识别领域中。通过研究分析,将学习向量量化(Learning Vector Quantization, LVQ)神经网络应用于雷达目标一维距离像识别。针对LVQ神经网络对初始连接权值敏感的问题和如何增强网络的分类识别性能,提出利用粒子群优化(Particle Swarm Optimization, PSO)算法对其进行优化。在此基础上提出了基于PSO-LVQ神经网络的雷达目标一维距离像识别新方法。通过3类飞机实测数据实验,验证了PSO算法优化LVQ神经网络初始连接权值的可行性和PSO-LVQ识别算法的有效性。 |
关键词: LVQ神经网络 粒子群算法 一维距离像 目标识别 |
DOI:DOI: 10.3969/j.issn.1672-2337.2019.01.006 |
分类号:TN957.52 |
基金项目:国家自然科学基金(No.61471379,61790551,61102166); 国防科技项目基金(No.2102028); 装备发展部“十三五”预研项目 (No.41413060101); 泰山学者工程专项经费资助 |
|
Radar Target Recognition Algorithm Based on Optimized LVQ Network |
GUO Xiaokang, JIAN Tao, DONG Yunlong
|
Research Institute of Information Fusion, Naval Aviation University, Yantai 264001, China
|
Abstract: |
The radar target one-dimensional range profile has high research value in radar target recognition field. The neural network has strong adaptability and is widely used in the field of target recognition. Through research and analysis,this paper applies learning vector quantization (LVQ) neural network to radar target 1-D range profile recognition. Based on the problem of LVQ neural network sensitive to initial connection weights and enhanced network classification recognition performance, this paper proposes to use particle swarm optimization (PSO) algorithm to optimize the weights. On this basis, a new method of radar one-dimensional range profile recognition based on PSO-LVQ neural network is proposed. Experimental results based on measured data of three kinds of aircraft verify the feasibility of PSO algorithm to optimize the initial connection weight of LVQ neural network and the effectiveness of PSO-LVQ algorithm. |
Key words: LVQ neural network particle swarm optimization one-dimensional range profile target recognition |