Title :
Minimum Norm Least Squares Extrapolation Estimate for Discrete (a,b,c,d)-Bandlimited Signals
Author :
Zhao, Hui ; Wang, Ruyan ; Song, Daiping ; Wu, Dapeng ; Zhang, Tianqi
Author_Institution :
Sch. of Commun. & Inf. Eng., Chongqing Univ. of Posts & Telecommun., Chongqing, China
Abstract :
This letter investigates the extrapolation problem of discrete (a,b,c,d)-bandlimited signals. Based on operator notations, we first show that the extrapolation problem amounts to finding an admissible solution of an under-determined system. Then the minimum norm least squares (MNLS) solution is shown to be an admissible extrapolation estimate. Besides, singular value decomposition method is used to set up an extrapolation formula which guarantees that the extrapolation result achieves the MNLS estimate. An iterative extrapolation algorithm is also proposed to achieve the MNLS estimate. Finally, numerical results are given to demonstrate the efficiency of the proposed extrapolation algorithms.
Keywords :
estimation theory; extrapolation; iterative methods; least squares approximations; signal processing; singular value decomposition; MNLS estimation; admissible extrapolation estimation; admissible solution; discrete (a,b,c,d)-bandlimited signals; iterative extrapolation algorithm; minimum norm least squares extrapolation estimation; operator notations; singular value decomposition method; under-determined system; Adaptive optics; Educational institutions; Extrapolation; Least squares approximation; Optical imaging; Singular value decomposition; Vectors; $(a,b,c,d)$-bandlimited; discrete signals; extrapolation estimate; minimum norm least squares solution;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2012.2223816