Title :
A rapid algorithm for on-line and real-time ARMA modeling
Author :
Shufang, Zhang ; Renjie, Liu
Author_Institution :
Coll. of Inf. Eng., Dalian Maritime Univ., China
Abstract :
We introduce a new practical method of on-line and real-time ARMA modeling. In this method a very high order AR model is used to approach the ARMA model. The multiplication of multi lower order AR models are used to replace the high order AR model, and the cascading linear systems are used to resolve the multiplication of the AR models. The lower order and linear algorithm can achieve the on-line and real-time resolution. The modeling experiments for some systems have proved that the method is effective, and the prediction based on the method also has higher precision. This method has been used to improve GPS positioning precision to achieve some better results
Keywords :
Global Positioning System; autoregressive moving average processes; modelling; parameter estimation; prediction theory; signal processing; ARMA model; GPS positioning precision; MARMA model; cascading linear systems; fast algorithm; high order AR model; on-line ARMA modeling; real-time ARMA modeling; Autocorrelation; Biology; Educational institutions; Global Positioning System; Least squares methods; Linear systems; Nonlinear equations; Parameter estimation; Physics; Predictive models;
Conference_Titel :
Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-5747-7
DOI :
10.1109/ICOSP.2000.894481