parametric time-frequency representation
[tfr,t,f] = tfrparam(x) [tfr,t,f] = tfrparam(x, t) [tfr,t,f] = tfrparam(x, t, N) [tfr,t,f] = tfrparam(x, t, N, p) [tfr,t,f] = tfrparam(x, t, N, p, L) [tfr,t,f] = tfrparam(x, t, N, p, L, ptrace) [tfr,t,f] = tfrparam(x, t, N, p, L, ptrace, Rxtype) [tfr,t,f] = tfrparam(x, t, N, p, L, ptrace, Rxtype, method) [tfr,t,f] = tfrparam(x, t, N, p, L, ptrace, Rxtype, method, q) [tfr,t,f] = tfrparam(...,"plot")
a vector of complex numbers of size Nx: the signal
a vector with integer values in [1 Nx]: time instant(s) (default : 1:Nx).
a positive integer: the number of frequency bins (default : max(256,Nx) ).
a positive integer: the last autocorrelation lag (default : 2 ).
an odd positive integer: the length of the window around the analyzed time sample (default : max(51,round(Nx/10))).
character string with possible values 'hermitian', 'fbhermitian', 'burg' or 'fbburg': the choice of the correlation matrix algorithm (default : 'fbhermitian')
Case does not matter.
a character string: can be either 'ar', 'periodogram', 'capon', 'capnorm', 'lagunas', or 'genlag'.
Case does not matter.
parameter for the generalized Lagunas method.
if one input parameter is 'plot', tfrqview is called and the time-frequency representation will be plotted.
x = fmconst(256,0.1) + fmlin(256,0.15,0.30) + fmlin(256,0.2,0.45); clf; gcf().color_map = jetcolormap(128); subplot(221) [tfr,t,f] = tfrparam(x,1:256,256,3,21,%f,'fbhermitian','ar'); grayplot(t,f,tfr'); title("AR") subplot(222) [tfr,t,f] = tfrparam(x,1:256,256,3,21,%f,'fbhermitian','periodogram'); grayplot(t,f,tfr'); title("Periodogram") subplot(223) [tfr,t,f] = tfrparam(x,1:256,256,3,21,%f,'fbhermitian','capon'); grayplot(t,f,tfr'); title("Capon") subplot(224) [tfr,t,f] = tfrparam(x,1:256,256,3,21,%f,'fbhermitian','lagunas'); grayplot(t,f,tfr'); title("Lagunas") | ![]() | ![]() |