A Novel Adaptive Nonlinear Filter-Based Pipelined Feedforward Second-Order Volterra Architecture

Si-Chuan Province Key Lab. of Signal & Inf. Process., Southwest Jiaotong Univ., Chengdu
IEEE Transactions on Signal Processing (Impact Factor: 2.79). 02/2009; 57(1):237 - 246. DOI: 10.1109/TSP.2008.2007105
Source: IEEE Xplore


Due to the computational complexity of the Volterra filter, there are limitations on the implementation in practice. In this paper, a novel adaptive joint process filter using pipelined feedforward second-order Volterra architecture (JPPSOV) to reduce the computational burdens of the Volterra filter is proposed. The proposed architecture consists of two subsections: nonlinear subsection performing a nonlinear mapping from the input space to an intermediate space by the feedforward second-order Volterra (SOV), and a linear combiner performing a linear mapping from the intermediate space to the output space. The corresponding adaptive algorithms are deduced for the nonlinear and linear combiner subsections, respectively. Moreover, the analysis of theory shows that these adaptive algorithms based on the pipelined architecture are stable and convergence under a certain condition. To evaluate the performance of the JPPSOV, a series of simulation experiments are presented including nonlinear system identification and predicting of speech signals. Compared with the conventional SOV filter, adaptive JPPSOV filter exhibits a litter better convergence performance with less computational burden in terms of convergence speed and steady-state error.

6 Reads
  • Source
    • "To solve the computational complexity problem, some truncated Volterra filters were proposed, such as second-order Volterra (SOV) [9] and thirdorder Volterra (TOV) [10]. The truncated Volterra filters have been widely applied to estimate and identify many nonlinear dynamic systems [11] [12] [13] [14], equalization and compensation of channel systems [15], image processing [16] and speech processing [17] [18]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Although the least mean pth power (LMP) and normalized LMP (NLMP) algorithms of adaptive Volterra filters outperform the conventional least mean square (LMS) algorithm in the presence of α-stable noise, they still exhibit slow convergence and high steady-state kernel error in nonlinear system identification. To overcome these limitations, an enhanced recursive least mean pth power algorithm with logarithmic transformation (RLogLMP) is proposed in this paper. The proposed algorithm is adjusted to minimize the new cost function with the p-norm logarithmic transformation of the error signal. The logarithmic transformation, which can diminish the significance of outliers under α-stable noise environment, increases the robustness of the proposed algorithm and reduces the steady-state kernel error. Moreover, the proposed method improves the convergence rate by the enhanced recursive scheme. Finally, simulation results demonstrate that the proposed algorithm is superior to the LMP, NLMP, normalized least mean absolute deviation (NLMAD), recursive least squares (RLS) and nonlinear iteratively reweighted least squares (NIRLS) algorithms in terms of convergence rate and steady-state kernel error.
    Full-text · Article · Aug 2015 · Digital Signal Processing
  • [Show abstract] [Hide abstract]
    ABSTRACT: Electroencephalogram (EEG) recordings often experience interference by different kinds of noise, including white, muscle, and baseline, severely limiting its utility. The recent research has demonstrated that discrete-time Volterra models can be successfully applied to reduce the broadband and narrowband noise. Their usefulness is mainly because of their ability to approximate to an arbitrary precision any fading memory system and their property of linearity with respect to parameters, the kernels coefficients. The main drawback of these models is their parametric complexity implying the need to estimate a huge number of parameters. Numerical results show that the developed algorithm achieves performance improvement over the standard filtered algorithm. This paper presents a Volterra filter (VF) algorithm based on a multichannel structure for noise reduction. Several methods have been developed, but the VF appears to be the most effective for reducing muscle and baseline noise, especially when the contamination is greater in amplitude than the brain signal. The present study introduces a new method of reducing noise in EEG signals in one step with low EEG distortion and high noise reduction. Applications with different real and synthetic signals are discussed, showing the validity of the proposed method.
    No preview · Article · Feb 2012 · Circuits Systems and Signal Processing
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we describe a technique to smooth the Q-switched pulse output from the large core fibre by effectively reducing the degree of coherence of the laser beam before launch and thus reducing contrast in the speckle. Whilst developed for PIV, the technique is capable of general applicability.
    No preview · Article · Jan 2005
Show more