Adaptive mixed-norm filtering algorithm based on S alpha SG noise model

作者：Zha Daifeng^*; Qiu Tianshuang

来源：Digital Signal Processing, 2007, 17(2): 475-484.

DOI：10.1016/j.dsp.2006.01.002

• adaptive filtering
• alpha-stable distribution
• second-order statistics
• fractional lower-order statistics
• impulsive noise
• mixed norm
• normalized step-size

The standard mixed-norm filtering algorithm exhibits slow convergence in stable distribution environment, requires a stationary operating environment, and employs a constant step-size that needs to be determined a priori. We proposed a new adaptive mixed moments filtering algorithm based on S alpha SG (symmetry alpha-stable Gaussian) noise model. The simulation experiments show that the proposed algorithm exhibits increased convergence rate and stability performance than the conventional mixed-norm algorithm.

• 出版日期2007-3
• 单位大连理工大学