From: <car...@us...> - 2012-04-13 09:28:36
|
Revision: 10206 http://octave.svn.sourceforge.net/octave/?rev=10206&view=rev Author: carandraug Date: 2012-04-13 09:28:27 +0000 (Fri, 13 Apr 2012) Log Message: ----------- convenc: help text to texinfo and basic input checking Modified Paths: -------------- trunk/octave-forge/main/comm/inst/convenc.m Modified: trunk/octave-forge/main/comm/inst/convenc.m =================================================================== --- trunk/octave-forge/main/comm/inst/convenc.m 2012-04-13 09:08:01 UTC (rev 10205) +++ trunk/octave-forge/main/comm/inst/convenc.m 2012-04-13 09:28:27 UTC (rev 10206) @@ -13,38 +13,53 @@ ## You should have received a copy of the GNU General Public License along with ## this program; if not, see <http://www.gnu.org/licenses/>. -% -- Function File: convenv(m, G, k) -% Compute the output of an (n, k, L) convolutional encoder with vector input -% m and matrix of generator polynomials G. -% -% The input vector m can be of arbitrary length. G is a matrix with n rows -% and k*(L+1) columns. The rows of G are the generator polynomials for each -% of the n output bits (per k input bits). -% -% The output is a vector whose length is n*floor([length(m)+k*(L+1)-1]/k). -% With two inputs, k is assumed to be equal to 1. -% -% Example 1: Compute the output from a (2, 1, 2) convolutional encoder -% with generator polynomials g1 = [ 1 1 1 ] and g2 = [ 1 0 1 ] -% when the input message is m = [ 1 1 0 1 1 1 0 0 1 0 0 0 ] -% -% x = convenc(m, [g1; g2]) -% x = 1101010001100111111011000000 -% -% Example 2: Compute the output from a (3, 2, 1) conv encoder with -% generator polynomials g1 = [ 1 0 1 1 ], g2 = [ 1 1 0 1 ] and -% g3 = [ 1 0 1 0 ] when the input is m = [ 0 1 1 0 0 0 1 1 ] -% -% x = convenc(m, [g1; g2; g3], 2) -% x = 111 111 110 101 -% -% Note: This function is not compatible with Matlab's convenc() +## -*- texinfo -*- +## @deftypefn {Function File} {@var{x} =} convenc (@var{m}, @var{G}, @var{k}) +## Compute output of an (n, @var{k}, L) convolutional encoder with vector input +## @var{m} and matrix of generator polynomials @var{G}. +## +## The input vector @var{m} can be of arbitrary length. @var{G} is a matrix with n rows +## and @var{k}*(L+1) columns. The rows of @var{G} are the generator polynomials for each +## of the n output bits (per @var{k} input bits). +## +## The output is a vector whose length is n*floor([length(@var{m})+@var{k}*(L+1)-1]/@var{k}). +## If unspecified, @var{k} defaults to 1. +## +## Example 1: Compute the output from a (2, 1, 2) convolutional encoder +## @example +## @group +## m = [ 1 1 0 1 1 1 0 0 1 0 0 0]; +## g1 = [1 1 1]; +## g2 = [1 0 1]; +## convenc (m, [g1; g2]) +## @result{} [1 1 0 1 0 1 0 0 0 1 1 0 0 1 1 1 1 1 1 0 1 1 0 0 0 0 0 0] +## @end group +## @end example +## +## Example 2: Compute the output from a (3, 2, 1) convolutional encoder +## @example +## @group +## m = [0 1 1 0 0 0 1 1 ]; +## g1 = [1 0 1 1]; +## g2 = [1 1 0 1]; +## g3 = [1 0 1 0]; +## convenc (m, [g1; g2; g3], 2) +## @result{} [1 1 1 1 1 1 1 1 0 1 0 1] +## @end group +## @end example +## +## @strong{Caution:}: this function is not compatible with @sc{matlab}'s convenc(). +## @end deftypefn -function x = convenc(m, G, k=1) - % Use conv2 to do repeated 1d convolutions of m with each row of G. - % rem is used to transform the standard convolution result to one - % which uses modulo-2 addition. Only cols with index a mult. of k - % are in the actual enc. output +function x = convenc (m, G, k = 1) + if (nargin < 2 || nargin > 3) + print_usage; + endif - x = rem(conv2(1, m(:)', G),2)(:,!rem(1:numel(m),k))(:)'; -end + # Use conv2 to do repeated 1d convolutions of m with each row of G. + # rem is used to transform the standard convolution result to one + # which uses modulo-2 addition. Only cols with index a mult. of k + # are in the actual enc. output + + x = rem(conv2(1, m(:)', G),2)(:,!rem(1:numel(m),k))(:)'; +endfunction This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site. |