System Identification Toolbox™

bode - Compute and plot frequency response magnitude and phase for logarithmic frequencies

Syntax

bode(m)
bode(m,w)
bode(m('noise')
bode(m1,...,mN,'sd',sd,'mode','same','ap',ap,'fill')
[mag,phase,w] = bode(m)
[mag,phase,w,sdmag,sdphase] = bode(m)

Description

bode(m) plots a Bode plot for the model m, which can be an idpoly, idss, idarx, idgrey, or idfrd object. This frequency response is a function of logarithmic frequencies in radians per unit time (stored as the TimeUnit model property). Default frequency values are computed from the model dynamics. For time series spectra, phase plots are omitted. For MIMO models, press Enter to view the next plot in the sequence of different I/O channel pairs, annotated using the InputNames and OuputNames model properties.

bode(m,w) plots a Bode plot at specified frequencies w in radians per unit time, which can be

• A vector of values.

• {wmin,wmax}, which specifies 100 logarithmically spaced frequency values ranging from a minimum value wmin and a maximum value wmax.

• {wmin,wmax,np}, which specifies np logarithmically spaced frequency values.

Note   For idfrd models, you cannot specify individual frequencies and can only limit the frequencies range for the internally stored frequencies using {wmin,wmax}.

bode(m('noise') plots a Bode plot of the output noise spectra when the model contains noise spectrum information.

bode(m1,...,mN,'sd',sd,'mode','same','ap',ap,'fill') plots a Bode plot for several models. sd specifies the confidence region as a positive number that represents the number of standard deviations. The argument 'fill' indicates that the confidence region is color filled. mode = 'same' displays all I/O channels in the same plot. Set ap = 'A' to show only amplitude plots, or ap = 'P' to show only phase plots.

[mag,phase,w] = bode(m) computes the magnitude mag and phase values of the frequency response, which are 3-D arrays with dimensions (number of outputs)-by-(number of inputs)-by-(length of w). w specifies the frequency values for computing the response even if you did not specify it as an input. For SISO systems, mag(1,1,k) and phase(1,1,k) are the magnitude and phase (in degrees) at the frequency w(k). For MIMO systems, mag(i,j,k) is the magnitude of the frequency response at frequency w(k) from input j to output i, and similarly for phase(i,j,k). When m is a time series, mag is its power spectrum and phase is zero.

[mag,phase,w,sdmag,sdphase] = bode(m) computes the standard deviations of the magnitude sdmag and the phase sdphase. sdmag is an array of the same size as mag, and sdphase is an array of the same size as phase.

See Also

bj

compare