Robust Geman-McClure Based Nonlinear Spline Adaptive Filter Against Impulsive Noise
Qianqian Liu, Yigang He
Abstract
This paper first proposes nonlinear spline adaptive filter based on the robust Geman-McClure estimator (SAF-RGM). The proposed algorithm is obtained by minimizing the cost function relied on the Geman-McClure estimator. Since the Geman-McClure estimator can remove outliers with large amplitude from dataset, the proposed algorithm can obtain the excellent performance in the impulsive noise. Moreover, the mean and mean square behaviors of the SAF-RGM algorithm are analyzed. Simulations are conducted to confirm that the proposed SAF-RGM algorithm achieves better performance than the existing spline nonlinear adaptive filtering algorithms. Besides, simulation results validate the theoretical conclusions.
Topics & Concepts
EstimatorSpline (mechanical)OutlierComputer scienceAlgorithmNonlinear systemNoise (video)Filter (signal processing)Adaptive filterMean squared errorMathematical optimizationMathematicsArtificial intelligenceImage (mathematics)PhysicsStatisticsComputer visionThermodynamicsQuantum mechanicsAdvanced Adaptive Filtering TechniquesBlind Source Separation TechniquesControl Systems and Identification