Enhanced Nonlinear System Identification by Interpolating Low-Rank Tensors
Authors:
- Christina AuerCAChristina Auer
- Thomas PairederTPThomas Paireder
- Oliver PloderOPOliver Ploder
- Oliver LangOLOliver Lang
- Mario HuemerMHMario Huemer
Preprint
Abstract
Function approximation from input and output data is one of the most investigated problems in signal processing. This problem has been tackled with various signal processing and machine learning methods. Although tensors have a rich history upon numerous disciplines, tensor-based estimation has recently become of particular interest in system identification. In this paper we focus on the problem of adaptive nonlinear system identification solved with interpolated tensor methods. We introduce three novel approaches where we combine the existing tensor-based estimation techniques with multidimensional linear interpolation. To keep the reduced complexity, we stick to the concept where the algorithms employ a Wiener or Hammerstein structure and the tensors are combined with the well-known LMS algorithm. The update of the tensor is based on a stochastic gradient decent concept. Moreover, an appropriate step size normalization for the update of the tensors and the LMS supports the convergence. Finally, in several experiments we show that the proposed algorithms almost always clearly outperform the state-of-the-art methods with lower or comparable complexity.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Nonlinear system identification with regularized Tensor Network B-splinesAutomatica, 2020
- Matrix output extension of the tensor network Kalman filter with an application in MIMO Volterra system identificationAutomatica, 2018
- A survey of deep neural network architectures and their applicationsNeurocomputing, 2017
- Nonlinear spline adaptive filteringSignal Processing, 2013
- Hilbert Space Methods in Signal Processing2013
- Blind identification of multiuser nonlinear channels using tensor decomposition and precodingSignal Processing, 2009
- Parallel factor analysis in sensor array processingIEEE Transactions on Signal Processing, 2000
- Adaptive parallel-cascade truncated Volterra filtersIEEE Transactions on Signal Processing, 1998
- Support-vector networksMachine Learning, 1995
- Nonlinear Problems in Random TheoryPhysics Today, 1959