摘要
提出了一种基于分形布朗运动模型的S波段雷达海杂波分形维数提取方法.分析了基于记忆库混沌时间序列预测方法,引入一种改进核函数的支持向量机分类器.在此基础上,提出了一种新的海杂波背景下目标检测方法.应用S波段雷达实测海杂波数据,计算得到了该信号的分形维数与Lyapunov指数,验证了S波段雷达海杂波的混沌分形特性.仿真实验结果验证了该方法具有较强的检测能力和抗杂波性能.
Adopting the model of fractional Brownian motion, this paper presents the method of deducing Hurst exponent based on the observed sea clutter of S-band radar. Secondly the prediction technology of chaotic time series is studied based on memory-based predictor. Furthermore, adopting the method of support vector machine classifiers of the improved radial basis kernel function, this paper proposes a novel method of target detection based on the sea clutter. Thirdly, on the basis of observed sea clutter of Sband radar, the fractal dimension and the largest Lyapunov exponent are obtained, which proves its chaos and fractal characteristic. Finally, the computer simulation is carried out and the results prove the effective detection performance and noise tolerance.
出处
《物理学报》
SCIE
EI
CAS
CSCD
北大核心
2006年第8期3985-3991,共7页
Acta Physica Sinica
基金
国家自然科学基金(批准号60402032)资助的课题.~~
关键词
分形布朗运动
分形维数
记忆库预测方法
支持向量机分类器
fractional Brownian motion, fractal dimension, memory-based predictor, support vector machine classifiers