摘要
针对海杂波中的弱信号检测问题,以相空间重构和模糊理论为基础,提出了一种基于T-S模型的模糊聚类方法对混沌时间序列进行预测和目标检测,利用自适应门限法判决混沌背景下微弱目标信号的有无。在模糊聚类建模中将前件划分和结论参数分开辨识,既简化辨识步骤,又提高模型的泛化能力,同时解决了模糊模型随辨识系统复杂程度提高而规则数增大的问题。与基于RBF神经网络的混沌背景下弱信号检测结果进行比较,仿真结果验证了该方法的有效性。
A method of chaotic time series prediction and target detection based on T-S fuzzy clustering model is proposed, which is based on the theory of phase-space reconstruetion and fuzzy theory. Chaotic time series is predicted and weak target signal is detected using the fuzzy clustering model, and the target signal is determined with adaptive threshold decision. In T-S fuzzy model the premise and conclusion are identified separately, which simplifies the steps of the identification and also improves the generalization ability. Meanwhile, the problem that the number of rules increases as the complicated degree of the identification system increases is also solved. Compared with RBF network on the weak signal detection under fuzzy background, simulation results demonstrate the effectiveness of the proposed method.
出处
《电子测量与仪器学报》
CSCD
2008年第5期53-58,共6页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金项目(50677014)
高校博士点基金项目(20060532016)
教育部新世纪优秀人才支持计划(NCET-04-0767)
湖南省自然科学基金项目(06JJ2024)
关键词
海杂波
混沌时间序列预测
模糊聚类
目标检测
sea clutter, chaotic time series prediction, fuzzy clustering, target detection.