function mixer = hohmann2002_mixer(fb, delay, iterations)
%hohmann2002_mixer Create new mixer object within HOHMANN2002 filterbank framework
% Usage: mixer = hohmann2002_mixer(fb, delay)
% mixer = hohmann2002_mixer(fb, delay, iterations)
%
% Input parameters:
% fb : A filterbank structure as created by |hohmann2002|
% The mixer created by this function can
% act as part of a synthesizer that resynthesizes the output
% from the input signal analyzed by `fb`.
% delay : A delay structure as created by |hohmann2002_delay|, together
% with the mixer created by this function, this delay can
% form a synthesizer.
% iterations : The gain factors are approximated numerically in
% iterations. If this parameter is omitted, then the
% number of iterations from `fb` will be used.
%
% `hohmann2002_mixer` creates a mixer object with gain factors suitable
% to calculate a weighted sum of the bands present in the output of the
% given delay. The gain factors are computed using a numerical optimization
% method described in Herzke & Hohmann (2007).
% author: Universitaet Oldenburg, tp (Jan 2002, Jan, Sep 2003, Nov 2006, Jan 2007)
% Adapted to AMT (PM, Jan 2016) from function gfb_mixer_new
if (nargin < 4)
iterations = fb.gaincalc_iterations;
end
mixer.type = 'gfb_mixer';
center_frequencies = fb.center_frequencies_hz;
number_of_bands = length(center_frequencies);
sampling_frequency = fb.fs;
% The center frequencies in the z plain
z_c = exp(2i * pi * center_frequencies(:) / sampling_frequency);
mixer.gains = ones(number_of_bands, 1);
% compute the frequency response of each filter (col) at the center
% frequencies of all filters (row)
pos_f = hohmann2002_freqz(fb, z_c);
neg_f = hohmann2002_freqz(fb, conj(z_c));
% apply delay and phase correction
for band = 1:number_of_bands
pos_f(:,band) = pos_f(:,band) * delay.phase_factors(band) .* z_c .^ -delay.delays_samples(band);
neg_f(:,band) = neg_f(:,band) * delay.phase_factors(band) .* conj(z_c) .^ -delay.delays_samples(band);
end
% combine responses at positive and negative responses to yield
% responses for real part.
f_response = (pos_f + conj(neg_f)) / 2;
for i = 1:iterations
% add selected spectrum of all bands with gain factors
selected_spectrum = f_response * mixer.gains;
% calculate better gain factors from result
mixer.gains = mixer.gains ./ abs(selected_spectrum);
end
mixer.gains = mixer.gains.';