Sparsity-aware normalized subband adaptive filters with jointly optimized parameters

Sparsity-Aware Robust Normalized Subband Adaptive Filtering Algorithms With Alternating Optimization of Parameters

“…Remark 1: (18) clearly reveals that the proposed SA-RNSAF algorithm will outperform the RNSAF algorithm for identifying sparse systems, if and only if △(k) < 0 holds 2 . It follows that ρ(k) should satisfy the inequality…”

“…Remark 2: The proposed AOP-SA-RNSAF update generalizes different algorithms, depending on the choice of ϕ(e) in (4b) and f (w) in (3). In the literature, several robust criteria against impulsive noises [6], [7], [9], [10], [14] defined by ϕ(e) and sparsity-aware penalties [15], [16], [18], [19], [22] defined by f (w) have been studied, which can be applied in the AOP-SA-RNSAF. Nevertheless, this paper does not consider the effect of different choices of ϕ(e) and/or f (w), which is worth studying in future work.…”

“…In the literature, they are rarely discussed simultaneously regardless of the Gaussian noise or impulsive noise scenarios. In [18], the variable parameter SA-NSAF (VP-SA-NSAF) algorithm was proposed for the Gaussian noise, in which these two parameters are jointly adapted based on a model-driven method, but it requires knowledge of variances of the subband noises. In [19], by optimizing the parameters in the sparsity-aware individualweighting-factors-based sign subband adaptive filter (S-IWF-SSAF) algorithm with the robustness in the impulsive noise, the variable parameters S-IWF-SSAF (VP-S-IWF-SSAF) algorithm was presented, while it lacks the generality in sparsityaware subband algorithms.…”

Sparsity-Aware Robust Normalized Subband Adaptive Filtering algorithms based on Alternating Optimization

“…Remark 2: The proposed AOP-SA-RNSAF update generalizes different algorithms, depending on the choice of ϕ(e) in (4b) and f (w) in ( 3). In the literature, several robust criteria against impulsive noises [14], [9], [6], [7], [10] defined by ϕ(e) and sparsity-aware penalties [19], [16], [18], [15], [95] defined by f (w) have been studied, which can be applied in the AOP-SA-RNSAF. Nevertheless, this paper does not consider the effect of different choices of ϕ(e) and/or f (w), which is worth studying in future work.…”

A Novel Normalized Subband Adaptive Filter Algorithm Based on the Joint-Optimization Scheme

“…However, despite the lower computational complexity than the AP algorithm, the NSAF algorithms have better performance in terms of tracking speed owing to their self-whitening property. Moreover, some variants of the NSAF, such as sign subband adaptive filter (SSAF) and its variants [10]- [12], M-estimate NSAF [13], [14], sparsity-aware SSAF and NSAF [15], [16], and bias-compensated NSAF [17]- [19], have been proposed to improve performance.…”