function [J,Msp,Nsp]=mtfftpb(data,tapers,nfft)
% Multi-taper fourier transform - binned point process data
%
% Usage:
%
% [J,Msp,Nsp]=mtfftpb(data,tapers,nfft) - all arguments required
% Input: 
%       data   (in form samples x channels/trials or single vector) 
%       tapers (precalculated tapers from dpss)  
%       nfft   (length of padded data)
% Output:
%       J (fft in form frequency index x taper index x channels/trials)
%       Msp (number of spikes per sample in each channel)
%       Nsp (number of spikes in each channel)

if nargin < 3; error('Need all input arguments'); end;
data=change_row_to_column(data); % changes data stored as a row vector to a column vector
[N,C]=size(data); % size of data
K=size(tapers,2); % size of tapers
tapers=tapers(:,:,ones(1,C)); % add channel indices to tapers
H=fft(tapers,nfft,1); % fourier transform of the tapers
Nsp=sum(data,1); % number of spikes in each channel
Msp=Nsp'./N; % mean rate for each channel
meansp=Msp(:,ones(1,K),ones(1,size(H,1)));  % add taper and frequency indices to meansp
meansp=permute(meansp,[3,2,1]); % permute to get meansp with the same dimensions as H
data=data(:,:,ones(1,K));% add taper indices to the data
data=permute(data,[1 3 2]); % permute data to be of the same dimensions as H
data_proj=data.*tapers; % multiply data by the tapers
J=fft(data_proj,nfft,1); % fft of projected data
J=J-H.*meansp; % subtract the dc