一类非均匀采样系统最小二乘迭代辨识  

Least-square-iterative identification of a class of non-uniform sampled-data systems

作　　者：蒋红霞[1] 王金海[1] 丁锋[1] 

机构地区：[1]江南大学通信与信息工程研究中心,江苏无锡214122

出　　处：《系统工程与电子技术》2008年第8期1535-1539,共5页Systems Engineering and Electronics

基　　金：国家自然科学基金(60574051);江苏省自然科学基金(BK2007017);江南大学创新团队发展计划资助课题

摘　　要：阐述了非均匀多率采样方案,利用提升技术推导出非均匀多率采样系统的提升状态空间模型,进一步获得了对应的输入输出表达。针对辨识模型信息向量中存在未知中间变量的困难,基于交互估计理论与迭代搜索原理,提出了非均匀多率系统的最小二乘迭代辨识方法。最后进了仿真比较研究,以说明提出算法的优点。The sampling patterns of non-uniform sampling systems are described, and the lifted state space model for non-uniform multirate systems by using the lifting technique is derived. Further, the corresponding input output relationship is obtained. For difficulties arising from unmeasurable information vectors in the iden- tification model, a least-square-iterative identification method for the non-uniform multirate system based on the interactive estimation theory and iterative search principle is presented. Finally, a simulated example is given to show the merits of the algorithm.

关 键 词：多率系统 参数估计 迭代辨识 提升技术 状态空间模型 传递函数模型 

分 类 号：TP271[自动化与计算机技术—检测技术与自动化装置]