Ideal TFR for given instantaneous frequency laws
[TFR,T,F] = tfrideal(IFLAWS) [TFR,T,F] = tfrideal(IFLAWS, T,) [TFR,T,F] = tfrideal(IFLAWS, T, N) [TFR,T,F] = tfrideal(IFLAWS, T, N, TRACE) [TFR,T,F] = tfrideal(...,'plot')
(M,P)-matrix where each column corresponds to the instantaneous frequency law of an (M,1)-signal, These P signals do not need to be present at the same time (missing values are represented by NaN's. The values of IFLAWS must be between 0 and 0.5.
a vector of integer: the time instant(s) (default : 1:M).
a positive integer: the number of frequency bins (default : M).
if non zero or %t, the progression of the algorithm is shown (default : %f).
if one input parameter is 'plot', tfrideal runs tfrqview. and TFR will be plotted
a real N by length(T) matrix with values in {0 1 %nan}: the output time-frequency matrix.
a real vector: the normalized frequencies.
tfrideal generates the ideal time-frequency representation corresponding to the instantaneous frequency laws of the components of a signal.
N = 140; [x1,if1] = fmlin(N,0.05,0.3); [x2,if2] = fmsin(70,0.35,0.45,60); if2 = [zeros(35,1)*%nan ; if2 ; zeros(35,1)*%nan]; [tfr,t,f] = tfrideal([if1 if2],1:N-1); clf contour(t,f,tfr',1); xlabel(_('Time')); ylabel(_('Normalized frequency')); title(_('Ideal time-frequency representation')); | ![]() | ![]() |