摘要
研究了一种 AR模型的盲辨识算法 ,该算法可同时辨识 AR模型的阶次及相应的模型参数 ,而不需事先确定 AR模型的阶次或者假定 AR模型的阶次已知 .其特点是计算量小 ,具有很好的收敛性 ,适合在线建模 ,并与常用的奇异值 -最小二乘法相比较 ,无论是运行速度、还是辨识精度方面 ,都具有优势 ,尤其随着阶次的增加 ,优势越明显 .仿真结果表明该算法是很有效的 .
Blind ielentifitication algorithm of AR model was studied and showed that the algorithm could identify system order and the parameters of AR model simultaneously; and that the system order need not be known or determined beforehand. The distinguishing feature of the algorithm is that it needs limited amount calculation and has great adapt ability to modeling online; and that compared singular value decomposition method and Least square method (SVD LS), whether in speed, or in identification accuracy, the algorithm has distinct advantages, especially at higher system order. The simulation results showed that the algorithm is very efficient.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2002年第4期381-384,共4页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金资助项目 (5 0 0 75 0 79)
