SLICOT system identification software and applications

Basic algorithmic, numerical and practical issues involved in the recently developed, multivariable system identification toolbox-SLIDENT-incorporated in the Fortran 77 Subroutine Library in Control Theory (SLICOT) are described. SLIDENT provides drivers, computational routines, and Matlab or Scilab interfaces, which implement several algorithmic approaches, and use standard or fast techniques for data processing. Both linear and Wiener-type multivariable discrete-time systems are addressed. The results obtained through an extensive investigation of the performance of the SLICOT system identification toolbox on a representative collection of data sets show that SLIDENT is reliable and able to solve large identification problems.