function dat = ft_preproc_medianfilter(dat, order)

% FT_PREPROC_MEDIANFILTER applies a median filter, which smooths the data with a

% boxcar-like kernel, except that it keeps steps in the data. This function requires

% the MATLAB Signal Processing toolbox.

%

% Use as

% [dat] = ft_preproc_medianfilter(dat, order)

% where

% dat data matrix (Nchans X Ntime)

% order number, the length of the median filter kernel (default = 25)

%

% If the data contains NaNs, these are ignored for the computation, but

% retained in the output.

%

% See also PREPROC

% Copyright (C) 2008, Robert Oostenveld

%

% This file is part of FieldTrip, see http://www.fieldtriptoolbox.org

% for the documentation and details.

%

% FieldTrip is free software: you can redistribute it and/or modify

% it under the terms of the GNU General Public License as published by

% the Free Software Foundation, either version 3 of the License, or

% (at your option) any later version.

%

% FieldTrip is distributed in the hope that it will be useful,

% but WITHOUT ANY WARRANTY; without even the implied warranty of

% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

% GNU General Public License for more details.

%

% You should have received a copy of the GNU General Public License

% along with FieldTrip. If not, see <http://www.gnu.org/licenses/>.

%

% $Id$

% set the default filter order

if nargin<2 || isempty(order)

ft_error('the order of the median filter is not specified');

end

% preprocessing fails on channels that contain NaN

if any(isnan(dat(:)))

ft_warning('FieldTrip:dataContainsNaN', 'data contains NaN values');

end

% deal with padding

pad = ceil(order/2);

dat = ft_preproc_padding(dat, 'localmean', pad);

hasfast = exist('fastmedfilt1d', 'file');

if hasfast == 2 || hasfast == 3

% use fast median filter mex file

for k = 1:size(dat,1)

dat(k,:) = fastmedfilt1d(dat(k,:), order);

end

else

is_matlab=ft_platform_supports('matlabversion',1,inf);

if is_matlab

% use Mathworks slow version

dat = medfilt1(dat, order, [], 2);

else

% use helper function that uses Octave's medfilt1

dat = medfilt1_rowwise(dat, order);

end

end

% cut the eges

dat = ft_preproc_padding(dat, 'remove', pad);

%%%%%%%%%%%%%%%%%%%%%%

% Helper function

%%%%%%%%%%%%%%%%%%%%%%

function y = medfilt1_rowwise(x,order)

% this function is compatible with Octave;

% Octave's medfilt1 accepts only two input arguments

y = zeros(size(x));

for k = 1:size(x,1)

y(k,:) = medfilt1(x(k,:),order);

end