摘要
基于HongLong多速率模型,引入小波变换进行奇异值检测,快速剔除野值;利用子波分解抽取多速率信息用于状态估计,建立了抗野值多速率模型.结合交互式多模型算法,形成野值条件下的机动目标状态估计算法.仿真结果表明,该算法对观测中的野值有良好的抗干扰性,并且通过抽取观测序列包含的多速率信息实现了对目标状态的精确估计.
Based on the theory of multirate kinematic model exploited by Hong, wavelet transform was introduced to detect the outliers and eliminate the influence fleetly. A novel multirate kinematic model restraining outliers was presented, more reasonable information was extracted for state estimation. The new model is used with interacting multiplemodel(IMM) algorithm in state estimation. The simulation results prove that it can restrain the outliers effectively and increase the accuracy of estimation.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2005年第9期1500-1504,共5页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金资助项目(60304007)
上海市科技启明星资助计划(04QMX1410)
关键词
小波变换
野值剔除
多速率模型
交互式多模型
wavelet transform
outlier elimination
multirate model
interacting multiple mode (IMM)
作者简介
肖艳军(1981-),男,河南信阳人,硕士生,主要从事智能信号与信息处理研究。
李建勋(联系人),男,教授,博士生导师,电话(Tel,):021-63932119;E-mail:lijx@sjtu,edu.cn.