A generalized proportionate variable step-size algorithm for fast changing acoustic environments

A robust correntropy based subspace tracking algorithm in impulsive noise environments

“…Based on the measure of the variation of system state, which is first proposed in [21], the approximation of subspace variations is developed and adopted to guide the selection of the number of the measurements ( ) L t in updating the Kalman filter [5] [13]. In this paper, we achieve a robust selection of the FF in the MCC-PAST in a similar way.…”

Sparsity Regularized Affine Projection Adaptive Filtering for System Identification

“…Figure 2 illustrates the MSD curves of the classical APs and two SR-APs, i.e., SR-AP-1 and SR-AP-2, in the case of SNR = 30dB. For comparison purpose, the improved proportionate AP (IPAP) [6], [7] and variable step-size AP (VS-AP) [11] are considered. The parameters, α = −0.5 for the IPAP and C = 0.003 for the VS-AP, are chosen, respectively.…”

“…Examples include echo paths [4] and multipath wireless communication channels [5]. To address this issue, variants of the AP algorithm which employ the variable gain parameters in accordance with the magnitude of the filter weights have been presented [6], [7]. However, these proportionate AP algorithms do not exploit the sparsity condition of an underlying system to be identified.…”

A family of proportionate normalized subband adaptive filter algorithms

“…Whereas existing algorithms use the same relationship for all coefficients, in IIPNLMS, the algorithm chooses between two different relationships depending on whether the coefficient is considered active or inactive. The concept of proportionate adaptation was also successfully extended to affine projection (AP) algorithm in [18] and PAP algorithm was developed.…”