Convergence Analysis of RLS Algorithms under Weak Conditions

摘要：

This paper establishes the consistency of recursive least squares and extended least squares algorithms for identifying ARX and ARMAX models under very weak assumptions. We relax the assumptions that the noise variance is constant and that high-order moments are existent, and prove that the parameter estimates consistently converge to the true parameters under time-varying variance and unbounded condition number, and that the parameter estimation error converges to zero under bounded condition number and unbounded variance.

关键词： Recursive identification parameter estimation convergence properties least squares extended least squares

作者: Feng Ding Yi Zhou Ming Li Jiyang Dai

作者单位: Control Science and Engineering Research Center Southern Yangtze University Wuxi, Jiangsu, P.R. Chin Key Laboratory of Nondestructive Test Nanchang Institute of Aeronautical Technology Nanchang, Jiangx

会议类型: 国际会议

会议名称: 第三届国际脉冲动力系统及应用学术会议

会议地点: 青岛

会议语种:英文

在线出版日期: 2006-07-21（万方平台首次上网日期，不代表论文的发表时间）