Hi Bernard,
I understand that you want to decode the messages containing zeros using “arithenco” and “arithdeco” functions of MATLAB. You can modify the code to map the “decodedY” values back to the “source” as you have used “unique” function to map the “message” to “seq”. Here is the modified code:
clearvars;
close all;
message = [2,1,2,1,2,0,2];
%% Encoder
table_size = height(message);
rows = table_size(1);
encodedY = [];
for row = 1:rows
coeff_row = message(row,:);
% Calculate unique values
source = unique(coeff_row);
for i=1:length(source)
counts(i)=length(strfind(coeff_row,source(i)));
end
seq=zeros(1,length(coeff_row));
for i=1:length(coeff_row)
seq(i)=strfind(source,coeff_row(i));
end
% Get arithmetic code for the block
code = arithenco(seq, counts);
% Store the block's code
encodedY = [encodedY; {code}];
end
%% Decoder
decodedY = [];
for row = 1 : size(encodedY)
temp = cell2mat(encodedY(row,:));
dseq = arithdeco(temp,counts,length(seq));
decodedY = [decodedY;dseq];
end
%modified here
decoded_message = zeros(size(decodedY));
for i = 1:length(decodedY)
decoded_message(i) = source(decodedY(i));
end
disp("message")
disp(message)
disp("decoded message")
disp(decoded_message)
%%
This can properly encode and decode any message containing zeros.
I hope this helps!
Thank you,
Abhimenyu.
