A symbolic analysis method for detecting parameter redundancy in circuit models used in identification.
The central contribution of this repository is the function detectRedundancy.m
. This function performs symbolic analysis upon two vectors: theta
, a model's parameters, and kappa
, how the parameters appear in the model. The symbolic analysis determines how many of the parameters in theta
can be estimated given the contents of kappa
by checking the rank of the jacobian matrix.
This code is largely based on the paper "Determining the parametric structure of models" - D. J. Cole et. al., 2010, Mathematical Biosciences. The Maple code at D. Cole's site was also invaluable in development.
Much of the extended functionality of the Maple code has been excluded as the main objective of this function is only to detect the redundancy. The Maple code offers a method of determining what combinations can be estimated, but this requires a PDE solver which MATLAB does not have.