摘要
针对固定步长的最小二乘递推算法在信号辨识中采集数据量大、辨识速度较慢的缺点,提出了变步长的最小二乘递推算法,在保证同等辨识精度的条件下, 采集的数据明显减少,大大提高了辨识速度。
This article aims at the problem that RLS of the fixed step demand collecting the large quantity data and identifiably speed slow in the signal identification, bring upping RLS of changing step makes, under guarantee the condition that equal identification accuracy, Obvious decrease of data of collecting and Consumedly increased to the iden-tifiably speed, and obtain the good result. [
出处
《农机化研究》
北大核心
2004年第2期89-91,共3页
Journal of Agricultural Mechanization Research
关键词
变步长
最小二乘递推算法
信号辨识
模糊控制
auto-control technology
changed-step
theoretical research
fuzzy-logic control
least square algorithm