Coefficient Pruning for Higher-order Diagonals of Volterra Filters Representing Wiener-hammerstein Models
نویسندگان
چکیده
Block-based nonlinear structures as, e.g., the Wiener-Hammerstein model, are popular for analyzing a broad class of nonlinear distortions. On the other hand, these models can also be related to Volterra filters in diagonal coordinates which represent very general nonlinear filters with memory. In this work, we propose an approach for estimating the significant coefficients of the nonlinear Volterra kernels during adaptation of such block-based filters by relating the linear impulse response and the diagonals of higher-order kernels. As the number of coefficients for the higher-order kernels is generally very large, this information can then be used to prune the number of necessary coefficients and thus lower the computational complexity. Experimental results for several test signals and nonlinear systems demonstrate the effectiveness of such an approach in a nonlinear acoustic echo cancellation scenario.
منابع مشابه
Modeling Parallel Wiener-Hammerstein Systems Using Tensor Decomposition of Volterra Kernels
Providing flexibility and user-interpretability in nonlinear system identification can be achieved by means of block-oriented methods. One of such block-oriented system structures is the parallel WienerHammerstein system, which is a sum of Wiener-Hammerstein branches, consisting of static nonlinearities sandwiched between linear dynamical blocks. Parallel Wiener-Hammerstein models have more des...
متن کاملNonlinear System Identification Using Hammerstein-Wiener Neural Network and subspace algorithms
Neural networks are applicable in identification systems from input-output data. In this report, we analyze theHammerstein-Wiener models and identify them. TheHammerstein-Wiener systems are the simplest type of block orientednonlinear systems where the linear dynamic block issandwiched in between two static nonlinear blocks, whichappear in many engineering applications; the aim of nonlinearsyst...
متن کاملA reduced-rank approach for implementing higher-order Volterra filters
The use of Volterra filters in practical applications is often limited by their high computational burden. To cope with this problem, many strategies for implementing Volterra filters with reduced complexity have been proposed in the open literature. Some of these strategies are based on reduced-rank approaches obtained by defining a matrix of filter coefficients and applying the singular value...
متن کاملDevelopment of Nonlinear Black Box Models using Orthonormal Basis Filters: A Review*
Over the last two decades, there has been a growing interest in the use of orthonormal basis filters (OBF) for developing dynamic black box models. When compared with the conventional approach, a substantial dimensionality reduction can be achieved through OBF parameterization. Moreover, the orthonormal filters, because of their similarity to the Padé approximation, can model systems that exhib...
متن کاملIdentification of Hammerstein and Wiener Models Using Spectral Magnitude Matching
The problem of the identification of Hammerstein and Wiener models is considered in this paper. The suggested approach in this paper utilizes the spectral magnitude matching method that minimizes the sum squared error between the spectral magnitudes evaluated for a number of short-time frames of the measured output signal of the nonlinear system and the output signal of the nonlinear model. The...
متن کامل