source: liacs/NC2010/low-correlation/autocorrelation.m@ 169

Last change on this file since 169 was 46, checked in by Rick van der Zwet, 15 years ago

Report finished = sleep

  • Property svn:keywords set to Id
File size: 903 bytes
Line 
1% An objective function for the low-autocorrelation problem.
2% Author: Ofer M. Shir, 2004; oshir@liacs.nl.
3%-----------------------------------------------------------------------
4function [f] = autocorrelation(pop)
5% Given a population of binary sequences, this function calculates
6% the merit function according to the formula specified in the
7% exercise description. The input pop is the given matrix. The
8% output f is the merit factor calculated (row vector).
9
10n = size(pop,1);
11m = size(pop,2);
12E = zeros(1,m);
13
14% Calculated efficiently in a matrix-notation; auxilary matrices -
15% Y1,Y2 - are initialized in every iteration. They are shifted form
16% of the original y vectors. The diagonal of the dot-squared Y2*Y1
17% matrix is exactly the inner sum of merit function.
18for k=1:n-1
19 Y1=pop(1:n-k,:);
20 Y2=pop(k+1:n,:)';
21 E=E+((diag(Y2*Y1)).^2)';
22end
23
24% The output:
25f = (n*n*ones(1,m))./(2*E);
Note: See TracBrowser for help on using the repository browser.