| Products & Services | Solutions | Academia | Support | User Community | Company |
| Download Product Updates | | | Get Pricing | | | Trial Software |
| Documentation → System Identification Toolbox |
| Contents | Index |
| Learn more about System Identification Toolbox |
Datas = nkshift(Data,nk)
Data contains input-output data in the iddata format.
nk is a row vector with the same length as the number of input channels in Data.
Datas is an iddata object where the input channels in Data have been shifted according to nk. A positive value of nk(ku) means that input channel number ku is delayed nk(ku) samples.
nkshift supports both frequency- and time-domain
data. For frequency-domain data it multiplies with
to obtain the
same effect as shifting in the time domain. For continuous-time frequency-domain
data (Ts = 0), nk should
be interpreted as the shift in seconds.
nkshift lives in symbiosis with the InputDelay property of idmodel:
m1 = pem(dat,4,'InputDelay',nk)
is related to
m2 = pem(nkshift(dat,nk),4);
such that m1 and m2 are the same models, but m1 stores the delay information and uses this information when computing the frequency response, for example. When using m2, the delay value must be accounted for separately when computing time and frequency responses.
Note the difference from the idss and idpoly property nk.
m3 = pem(dat,4,'nk',nk)
gives a model that itself explicitly contains a delay of nk samples. In contrast, m1 contains a total delay of m1.nk + m1.InputDelay = 4.
| Algorithm Properties | |
| idss |
![]() | neuralnet | nlarx | ![]() |

Includes the most popular MATLAB recorded presentations with Q&A sessions led by MATLAB experts.
| © 1984-2009- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |