Title :
Learning curves for LMS and regular Gaussian processes
Author_Institution :
Coll. of Eng., Embry-Riddle Univ., Prescott, AZ, USA
fDate :
2/1/2002 12:00:00 AM
Abstract :
Uses methods due to Guo, Ljung, and Wang (1997) to obtain explicit bounds on the error of the LMS algorithm used in a linear prediction of a signal using previous values of that signal. The signal is assumed to be a mean-zero Gaussian regular stationary random process. The bounds are then used to construct learning curves for the LMS algorithm in situations where the statistics of the process are only partially known
Keywords :
Gaussian processes; least mean squares methods; matrix algebra; prediction theory; random processes; LMS; error bounds; learning curves; least mean squares process; linear prediction; mean-zero Gaussian regular stationary random process; Convergence; Covariance matrix; Gaussian processes; Least squares approximation; Linear matrix inequalities; Prediction algorithms; Random processes; Signal processing; Statistics; Stochastic processes;
Journal_Title :
Automatic Control, IEEE Transactions on