User`s guide

3 Linear Model Identification
To create a new model by subreferencing H due to unmeasured inputs, use
the following syntax:
m_H = m('noise')
Tip Altern a tively, you can use the followin g shortha nd synta x: m_H = m ('n' )
This operation creates a time-s erie s model from m by ignoring the m easured
input.
The covariance matrix of e is given by the
idmodel property N oise Variance,
which is the matrix
Λ
:
Λ=LL
T
The covariance m atrix of e is related to v,asfollows:
eLv=
where v is white noise with an identity covariance matrix representing
independent noise sources w ith unit variances.
Treating Noise Channels as Measured Inputs
To study noise contributio n s in more detail, it might be use fu l to convert the
noise channels to measured channels using
noisecnv:
m_GH = noisecnv(m)
This operation creates a model m_GH th at repr es en ts bo th m easured inputs u
and noise inputs e, treating both sources as measured signals. m_GH is a model
from
u and e to y, describing the transfer functions G and H.
3-122