基于小波去噪半参数回归模型的卫星轨道测量数据预处理方法_潘旺华.pdfVIP

　　 　　　　　　　 30 6 　　　　 　　JOURNAL OF NATIONAL UNIVERSITY OF DEFENSE T ECHNOLOGY　　　　 Vol.30 No.6 2008 :1001-2486 (2008)06-0010-04 1, 2 1 1 1 潘旺华 , 文援兰, 朱　俊, 廖　瑛 (1. , 　410073;　2., 　710025) 　:, 。 , , , 。 USB , : , , 。 :;;; :V412　　:A Study on Satellite Orbit Tracking Data Reprocessing with Semi-parametric Regression Model and Based on the Wavelet De-noising Approach 1,2 1 1 1 PAN Wang-hua , WEN Yuan-lan , ZHU Jun , LIAO Y ng (1.College of Aerospace and M ater al Eng neer ng, Nat onal Un v .of Def ense Technology, Changsha 410073, Ch na; 2.The Second Art llery Eng neer ng Inst tute, X an 710025, Ch na) Abstract:Trad t onally, the satell te orb t measurement data are reprocessed by least-square polynom al f t method.Becau se of the non-l near errors n the satell te track ng, the accuracy of data reprocess ng s reduced.The current approach, based on the sem - parametr c regress on model, holds that the observat ons can be de-no s ng v a w avelet threshold, and the nonl near errors can be est mated and removed from the observat ons.The data reprocess ng method w th sem -parametr c regress on model based on the wavelet de-nos ng approach s proposed to mprove the accuracy of data reprocess ng .F nally, the s mulat on of the reprocess ng of a un ted S-band (USB)satell te orb t measurement data show ed that the method could separate the w h te no se and nonl near errors, and greatly mprove the accuracy of data reprocess ng. Key words:least square method;polynom al f t;sem -parametr c regress on model;de-no s ng v a wavelet threshold , 。。,