|
samer@0
|
1 % additive_split - Arrow to process data with adaptive zero mean
|
|
samer@0
|
2 %
|
|
samer@0
|
3 % additive_split ::
|
|
samer@0
|
4 % arrow( {[N,M]}, {[N,M]}, S1) ~'arrow to process N channel zero mean data',
|
|
samer@0
|
5 % arrow( {[N]}, {[N]}, S2) ~'arrow to process estimated mean',
|
|
samer@0
|
6 % options {} ~'options passed to azeromean'
|
|
samer@0
|
7 % -> arrow( {[N,M]}, {[N,M]}, pair( pair([N],pair(S1,S2)), empty)).
|
|
samer@0
|
8 %
|
|
samer@0
|
9 % This arrow uses the azeromean arrow to adaptively estimate the mean of
|
|
samer@0
|
10 % the N channels of input data, using a gaussian data model. The mean is
|
|
samer@0
|
11 % subtracted out. The two components are then processed independently by
|
|
samer@0
|
12 % the two supplied arrows and added together to create the output.
|
|
samer@0
|
13
|
|
samer@0
|
14 function o=additive_split(a1,a2,varargin)
|
|
samer@37
|
15 opts=options(varargin{:});
|
|
samer@0
|
16
|
|
samer@0
|
17 o = states(azeromean(gaussian,opts))*(a1+a2)*aplus;
|
|
samer@0
|
18 end
|
|
samer@0
|
19
|
|
samer@0
|
20 function o=aplus, o=arrf(@plusfn,2,1); end
|
|
samer@0
|
21 function f=plusfn(sz)
|
|
samer@0
|
22 if all(sz{1}==sz{2}), f=@plus;
|
|
samer@0
|
23 elseif sz{1}(1)==sz{2}(1) && sz{1}(2)>sz{2}(2)
|
|
samer@0
|
24 f=@(a,b)a+repmat(b,1,size(a,2));
|
|
samer@0
|
25 elseif sz{1}(1)==sz{2}(1) && sz{1}(2)<sz{2}(2)
|
|
samer@0
|
26 f=@(a,b)b+repmat(a,1,size(b,2));
|
|
samer@0
|
27 else
|
|
samer@0
|
28 error('additive_split:incompatible sizes');
|
|
samer@0
|
29 end
|
|
samer@0
|
30 end
|
