“…In this section, simulation experiments will be setup to verify the performance of the KMEC algorithm over a NCE under different noise environments. In this paper, a nonlinear channel model consisting of a linear filter and a memoryless nonlinear model, and Gaussian kernel function in equation (3) is used in the simulation to model the NCE channel whose structure is shown in Fig. 1…”
In this paper, a novel kernel mixed error criterion (KMEC) algorithm is proposed for nonlinear system identification, which uses a combination of two different error schemes to implement a newly constructed cost function, which is realized by using a logarithmic squared error and a generalized maximum correntropy criterion (GMCC) to devise the KMEC algorithm. The proposed KMEC is derived in the context of the kernel adaptive filter and it provides good performance for identifying the nonlinear channels in different mixed noise environments in terms of the mean square error (MSE) at its steady-state and convergence performance. INDEX TERMS Kernel adaptive filtering, mixed error criterion algorithm, generalized maximum correntropy, non-Gaussian noise environments, nonlinear adaptive filtering.
“…In this section, simulation experiments will be setup to verify the performance of the KMEC algorithm over a NCE under different noise environments. In this paper, a nonlinear channel model consisting of a linear filter and a memoryless nonlinear model, and Gaussian kernel function in equation (3) is used in the simulation to model the NCE channel whose structure is shown in Fig. 1…”
In this paper, a novel kernel mixed error criterion (KMEC) algorithm is proposed for nonlinear system identification, which uses a combination of two different error schemes to implement a newly constructed cost function, which is realized by using a logarithmic squared error and a generalized maximum correntropy criterion (GMCC) to devise the KMEC algorithm. The proposed KMEC is derived in the context of the kernel adaptive filter and it provides good performance for identifying the nonlinear channels in different mixed noise environments in terms of the mean square error (MSE) at its steady-state and convergence performance. INDEX TERMS Kernel adaptive filtering, mixed error criterion algorithm, generalized maximum correntropy, non-Gaussian noise environments, nonlinear adaptive filtering.
“…At first glance from sparsity-aware strategies, compared with our previous work in [57], this work extends the additional proportionate mechanism for sparse systems. However, unlike [57], the PFBS-PNSAF algorithm is based on the PFBS and the soft-thresholding techniques.…”
Section: Introductionmentioning
confidence: 83%
“…At first glance from sparsity-aware strategies, compared with our previous work in [57], this work extends the additional proportionate mechanism for sparse systems. However, unlike [57], the PFBS-PNSAF algorithm is based on the PFBS and the soft-thresholding techniques. Importantly, this work also covers a comprehensive performance analysis for the PNSAF algorithm in terms of convergence condition, transient state and steady-state behaviors (which have not been discussed in detail).…”
Section: Introductionmentioning
confidence: 83%
“…In [56], the l 0 -norm penalty is considered into the NSAF algorithm, thereby obtaining a performance improvement when identifying sparse systems in the colored input scenarios. In [57], based on the l 1 -norm and reweighted l 1 -norm penalties, the sparsity-aware NSAF algorithms that outperform the NSAF algorithm were developed and analyzed. It is worth pointing out that the main role of the proportionate scheme is to speed up the convergence, while the sparsity-aware's role is to reduce the steady-state error.…”
In this paper, we propose a novel normalized subband adaptive filter algorithm suited for sparse scenarios, which combines the proportionate and sparsity-aware mechanisms. The proposed algorithm is derived based on the proximal forwardbackward splitting and the soft-thresholding methods. We analyze the mean and mean square behaviors of the algorithm, which is supported by simulations. In addition, an adaptive approach for the choice of the thresholding parameter in the proximal step is also proposed based on the minimization of the mean square deviation. Simulations in the contexts of system identification and acoustic echo cancellation verify the superiority of the proposed algorithm over its counterparts.
“…Moreover, the APA enjoys improved performance under the colored input signal [27]. In [28], a novel affine projection sign algorithm (APSA) was presented by taking use of the L 1 -norm algorithm for system identification in impulsive scenario. Following this work, several APSA-based algorithms were proposed by using variable step-size (VSS) scheme [29], [30], convex combination strategy [31] and so on.…”
In the field of signal processing such as system identification, the affine projection algorithm (APA) is extensively implemented. However, running such algorithms in a non-Gaussian scenario may degrade its performance, since the second-order moment cannot extract all information from the signal. To prevent performance degradation of the algorithm in system identification tasks, we propose a novel APA based on least mean fourth (LMF) algorithm. The new algorithm, namely affine projection least mean fourth algorithm (APLMFA) is based on the high-order error power (HOEP) criterion and as such, can achieve improved performance. We also provide a convergence analysis for APLMFA. Numerical simulation results verify the presented APLMFA achieves smaller steady-state error as compared with the state-of-theart algorithms. INDEX TERMS Affine projection algorithm, least mean fourth algorithm, high-order error power criterion, system identification.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.