TY - GEN
T1 - Adaptive FIR filtering under minimum error/input information criterion
AU - Chen, Badong
AU - Hu, Jinchun
AU - Li, Hong Bo
AU - Sun, Zengqi
PY - 2008
Y1 - 2008
N2 - In this paper, we use the mutual information between error/input as the cost function for adaptive filtering. For the finite-impulse response (FIR) filter, the connections between the minimum error/input information (MEII) criterion and traditional mean-square error (MSE) criterion are investigated. We show that, for Gaussian case, the MEII criterion is equivalent to the well-known orthogonality condition. Based on the MEII criterion and kernel density estimation, we derive a stochastic gradient algorithm. Simulation results emphasize the effectiveness of this new algorithm.
AB - In this paper, we use the mutual information between error/input as the cost function for adaptive filtering. For the finite-impulse response (FIR) filter, the connections between the minimum error/input information (MEII) criterion and traditional mean-square error (MSE) criterion are investigated. We show that, for Gaussian case, the MEII criterion is equivalent to the well-known orthogonality condition. Based on the MEII criterion and kernel density estimation, we derive a stochastic gradient algorithm. Simulation results emphasize the effectiveness of this new algorithm.
KW - Filtering and smoothing
KW - Iterative modelling and control design
KW - Recursive identification
UR - https://www.scopus.com/pages/publications/79961018999
U2 - 10.3182/20080706-5-KR-1001.1913
DO - 10.3182/20080706-5-KR-1001.1913
M3 - 会议稿件
AN - SCOPUS:79961018999
SN - 9783902661005
T3 - IFAC Proceedings Volumes (IFAC-PapersOnline)
BT - Proceedings of the 17th World Congress, International Federation of Automatic Control, IFAC
T2 - 17th World Congress, International Federation of Automatic Control, IFAC
Y2 - 6 July 2008 through 11 July 2008
ER -