DislikedHello folks from Noxa,
It sounds like you're talking about a principle components analysis along some set of feature vectors... probably implemented via SVD? Am I reading this right? If so, what's the general idea behind your feature vectors? You've said they're not simple sinusoids... but what are they?
If I've totally misinterpreted your post, please forgive me.
Cheers...Ignored
SSA looks at the trajectory matrix for repeating patterns representing oscillations. This can also be achieved by looking at the patterns that appear in the lagged covariance matrix through a decomposition into its set of eigenvalues and eigenvectors. And yes SVD can be used as the matrix is symetric.
Noxa