Title :
Volterra-TLS method for chaotic time series prediction
Author :
Li, Chunfu ; Yu, Juebang
Author_Institution :
Inst. of Astronaut. & Aeronaut., Univ. of Electron. Sci. & Technol. of China, Chengdu
Abstract :
A new Volterra-TLS method for chaotic time series prediction is proposed, which is based on Volterra series model, TLS criterion and iterative TLS algorithm. According to study results, the prediction precision of Volterra-TLS method is not inferior to that of Volterra-NLMS method, while Volterra-TLS method has higher training convergence speed and is more robust against noise.
Keywords :
Volterra series; chaos; least mean squares methods; signal processing; time series; Volterra series model; Volterra-total least square method; chaotic time series prediction; iterative algorithm; Chaos; Convergence; Finite impulse response filter; Iterative algorithms; Iterative methods; Least squares methods; Predictive models; Signal processing algorithms; Space technology; Vectors;
Conference_Titel :
Communications, Circuits and Systems, 2008. ICCCAS 2008. International Conference on
Conference_Location :
Fujian
Print_ISBN :
978-1-4244-2063-6
Electronic_ISBN :
978-1-4244-2064-3
DOI :
10.1109/ICCCAS.2008.4657725
