WSEAS Transactions on Signal Processing
Print ISSN: 1790-5052, E-ISSN: 2224-3488
Volume 15, 2019
Sparse System Identification in The Presence of Noisy Input Signal Using Biased Compensator Minimum Error Entropy Algorithm
Authors: ,
Abstract: Sparse systems are identified effectively by correntropy induced minimum error entropy (CIM-MEE) algorithm in the presence of non- Gaussian noise. But this does not take into account the noisy input signal. This paper presents a new approach for sparse system identification having input signal corrupted by white Gaussian noise. The noisy input signal produces a bias during estimation. The proposed scheme incorporates a bias compensator to overcome this bias by adding a constraint in objective function of CIM-MEE algorithm. Simulations carried out in MATLAB confirm better performance of proposed Biased Compensator CIM-MEE (BC-CIM-MEE) algorithm for noisy input signal in the presence of impulsive measurement noise.
Search Articles
Pages: 77-84
WSEAS Transactions on Signal Processing, ISSN / E-ISSN: 1790-5052 / 2224-3488, Volume 15, 2019, Art. #10