AN IDENTIFICATION TOOLBOX FOR PROFILING NOVEL TECHNIQUES

Abstract

This paper describes a Matlab (or Octave) based software package for the estimation of dynamic systems. It has been developed primarily as a vehicle for profiling novel approaches relative to existing methods within a common software framework that streamlines comparisons. Key features of the toolbox include simplicity of use (particularly via automated entry of unspecified values), the support of a wide range of scalar and multivariable model structures which include certain nonlinear classes such as bilinear and Hammerstein--Wiener, the ability to handle both time and frequency domain data, the hand optimisation of certain key routines compiled against ATLAS libraries for optimum speed, the use of non-standard optimisation methods based on adaptive subspace gradient search and the Expectation-Maximisation method, and the fact that the toolbox is freely available from http://sigpromu.org for non-commercial use.