Re: Prefered method for tone detection


This is just a suggestion. But a common form of detection (specially in
beamforming) is to obtain the SVD decomposition of the autocovariance
matrix of the data and create a detection paradigm based on the ratio
of the largest to the lowest eigenvalues (or the ration of the
geometric to arithmetic means of the eigenvalues).
If the ratio is high at least one signal is present. SInce you have
the autocovariance matrix already, you can use MUSIC to determine the
frequencies present (this might will require some computation though).



My filters are
2nd order direct form I filters, with the coefficents determined from a
elliptic filter. Is this a sensible method of detection?

.