Title :
ARMA Prediction of Widely Linear Systems by Using the Innovations Algorithm
Author :
Navarro-Moreno, Jesus
Author_Institution :
Dept. of Stat. & Oper. Res., Univ. of Jaen, Jaen
fDate :
7/1/2008 12:00:00 AM
Abstract :
A new autoregressive moving average (ARMA) predictor for widely linear systems is presented by using the innovations algorithm and under an improper treatment. The main characteristics of this predictor are its facility of implementation from a practical standpoint and a better performance with respect to the conventional predictor obtained from a proper processing.
Keywords :
autoregressive moving average processes; linear systems; signal processing; ARMA prediction; autoregressive moving average predictor; innovations algorithm; linear systems; signal processing; Autoregressive processes; Frequency estimation; Linear systems; MIMO; Optical signal processing; Power system modeling; Signal processing; Signal processing algorithms; Technological innovation; White noise; Autoregressive moving average (ARMA) models; improper complex random signals; innovations algorithm; linear least mean-square error (LLMSE) prediction problems; widely linear systems;
Journal_Title :
Signal Processing, IEEE Transactions on
DOI :
10.1109/TSP.2008.919396
