To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Adaptive Filtering of Non Stationary Signals for Frequency Estimation Using hybrid KDE |
|
Author |
B. N. Biswal, S. C. Satapathy, P .K. Patra, G.Panda |
| Citation |
Vol. 11 No. 6 pp. 288-294 |
|
Abstract |
The paper highlights a novel hybrid EKF (Extended Kalman Filter) and Differential evolution (DE) called KDE (Kalman differential evolution) approach for tracking Parameters of a signal corrupted with noise and harmonics. The modeling and measurement of error covariance matrices Q and R are optimized using differential evolution. A new adaptive filtering technique is employed for self adaption of mutation scale factor in DE. To improve the convergence speed of optimization process K-DE, a variant of standard DE is considered for the optimization of Q and R matrices. Simulation results for time varying frequency of the power signal reveal significant improvement in noise rejection and accuracy in obtaining the frequency of the signal. |
|
Keywords |
Differential evolution, Extended Kalman filters, covariance, Kalman Differential Evolution |
|
URL |
