nwidFreq_dcn function

Identify a full diffusively coupled network or a subnetwork from a diffusively coupled network using frequency domain identification.

Syntax

• model = nwidFreq_dcn(data,model,orders,options)  performs identification based on the provided data, initialized network model, model orders and option set. The identified network/subnetwork model is returned.

Input arguments

• data

  nwdata object

  Network data object. The input excitation, input node and output node signals (as referred to by their labels) in the Rset, Dset and Yset of the predictor model must be present in the data.

• model

  nwpredmodel object	nwmodel object

  Initialized network model object. Use an nwmodel object to identify the full network, and an nwpredmodel object to identify a subnetwork. The properties of the predictor mappings X, Y, B and F are used in the algorithm, with some restrictions:

  • At least 1 excitation is required to run estimation (i.e. B has at least 1 non-zero entry)
  • The excitation model B must be known
  • The noise model F is unused in this algorithm.
  • Since X, Y and B are polynomials, the number of denominator parameters npden must equal 0 for each element in these mappings.
  • For each element of X, it is assumed that the order (npnum) is at least 1 higher than the maximum order (npnum) of the corresponding row/column of Y.
  • Delay settings (d) are ignored.
  • All present elements of X and Y are assumed to be parametrized, the f property is ignored for these mappings.
  • Setting initial parameter values using pnum/pden is not supported.

• orders

  structure array

  Model orders. Can be used to override properties of the initialized network model object. Use a structure array with optional fields:

  • nA: Order of X elements, and order+1 of Y elements (nA = model.X.npnum-1 = model.Y.npnum)
  • nB: Order of B elements (nB = model.B.npnum-1)
  • nF: Order of F elements (nF = model.F.npnum-1)
  • nplm: Order of the local polynomial approximation used in the nonparametric estimation. Defaults to 6.
  • nig: Order of the transient term. Defaults to nA-1.
  • nAim: Order of X elements, and order+1 of Y elements modified by immersion.
  • nBim: Order of B elements modified by immersion.

  If the reduce_target option is true, the immersed network will be estimated with orders nAim and nBim, and the target subnetwork will then be estimated with order nA. All elements outside of the target subnetwork will be estimated with order nAim and nBim to account for the immersed dynamics, and the target subnetwork will be directly estimated with order nA.

• options

  nwidFreqOptions object

  Option set for nwidFreq_dcn function. Can be constructed using options = nwidFreqOptions for the default option set or options = nwidFreqOptions(Name=Value,...) to set specified options.

Output arguments

• model

  nwpredmodel object	nwmodel object

  Estimated network model object. This is an nwmodel object after full network identification, and an nwpredmodel object after subnetwork identification. The pnum and pden properties of the X, Y, B and F fields are populated with the estimated parameter values. The e properties of X, Y, B and F contain the fixed and estimated polynomials/transfer functions of the predictor model connections.