Reversing binary stream Conversion According to attached table?

Hi guys,

best regard for @dpb to help me to built this function in matlab:

function ret=binrestore(m,array)
% returns binary decoded array values from input encoded array
% multiple     Mapped Value = 0         Mapped Value= 1
%   1                     0                       1
%   2                    00                      11
%   4                  1100                    0011
%   8              11001100                00110011
assert(mod(numel(array),m)==0,'Input array length not multiple of multipler')
array=reshape(array,m,[]);        % separate out digits to compare
switch m
  case 1  % nothing else to do at this point
  case 2
    array=all(array==1);
  case 4
    array=all(array==[0 0 1 1].');
  case 8
    array=all(array==[0 0 1 1 0 0 1 1].');
  otherwise
    error('Multiplier not 1, 2, 4, 8')  
end
ret=num2str(array,'%d');
end  

this function does the binary converstion according to to the attached table, for example:

binrestore(1,[0 1]) ----> output= [0 1] . because 0 is mapped to zero according to table at m=1, and 1 is mapped to one.

binrestore(2,[0 1]) ----> output= [0 0 1 1 ]. because 0 is mapped to 0 0 and 1 is mapped to 1 1 according to table.

binrestore(4,[0 1]) ----> output= [1 1 0 0 0 0 1 1]. because 0 is mapped to 1 1 0 0 and 1 is mapped to 0 0 1 1 according to table.

what Im trying to implement is the reverse of this function, it means the output of this function is my input to the function that I want to implement which it's called reversebinrestore(m,array) and its output is the input to my function above.

for example of the function that Im trying to implement, If I input to it:

reversebinrestore(1,[0 1]) --> Output=[0 1]

reversebinrestore(2, [0 0 1 1 ]) --> Output=[0 1]

reversebinrestore(4,[1 1 0 0 0 0 1 1]) ---> Output=[0 1] , this because [1 1 0 0] according to table is mapped to zero, and [0 0 1 1 ] is mapped to one.

..etc

exactly the same functionality of binrestore(m ,array) but in opposite/reverse.

Any help please how I can I implement that in matlab? thanks alot!

[Jimmy cho]

Could be much appreciated if you could give me aid.

@Walter Roberson

Answer 1

[David Hill]

function ret=reversebinrestore(m,array)
  ret=[];
  switch m
      case 1
          l=[0;1];
          for k=1:length(array)
              ret=[ret,find(ismember(l,array(k)))-1];
          end
      case 2
          l=[0 0;1 1];
          array=reshape(array,2,[])';
          for k=1:size(array,1)
              ret=[ret,find(ismember(l,array(k,:),'rows'))-1];
          end
      case 4
          l=[1 1 0 0;0 0 1 1];
          array=reshape(array,4,[])';
          for k=1:size(array,1)
              ret=[ret,find(ismember(l,array(k,:),'rows'))-1];
          end
      case 8
          l=[1 1 0 0 1 1 0 0;0 0 1 1 0 0 1 1];
          array=reshape(array,8,[])';
          for k=1:size(array,1)
              ret=[ret,find(ismember(l,array(k,:),'rows'))-1];
          end
  end
end

[Jimmy cho]

Hi !

Thanks for your help but your code isn't giving me correct outputs.

I tried in my command matlab window:

>> reversebinrestore(2,[0 1])

ans =

  0×1 empty double column vector

which it must be the output= [0 0 1 1 ]

also I tried:

 reversebinrestore(4,[0 1])
Error using reshape
Product of known dimensions, 4, not divisible into total number of elements, 2.

Error in reversebinrestore (line 17)
          array=reshape(array,4,[])';
          

if I input array [0 1] with m=2 for instance, then 0 is mapped according to table to [0 0] and 1 is mapped to [1 1] so the output is [0 0 1 1]

if I input array [0 1 1] with m=4 for instance then, 0 is mapped according to table to [1 1 0 0] and 1 is mapped to [0 0 1 1 ]

so the output is [1 1 0 0 0 0 1 1 0 0 1 1 0 0 1 1 ];

[David Hill]

You are inputting the wrong thing. Your inputs are not in accordance with your description. It works perfectly for me if you input the correct thing.

reversebinrestore(2, [0 0 1 1 ]);
reversebinrestore(4,[1 1 0 0 0 0 1 1]);
reversebinrestore(2,[0 0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 1 1 1 1 1 1 0 0]);
reversebinrestore(8, [0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 ]);

[Jimmy cho]

Hi, I appreciate your effort , but you didn't understand me, the function above (attached code) is exactly I've done it and it's using the same thing as what you've done.

what Im trying to do is exactly the opposite, it means if I input [0 1] with m=1 then output is [0 1]

if I input [0 1] and m=2 then output is [0 0 1 1 ] ..etc

exactly that's what I called it "reverse" of my function above(reverse of what my attached code above does..)

Could you please help me how to implement the reverse function in matlab? once again sorry for the missunderstanding in my thread, and exactly what you're inputting is my output and the output of what you've coded is my input (the compatible binary stream).

Once again, I want to implement the function that it does the same thing as what you've done but in reverse, so if I input [0 1] with m=2, reversebinrestore(2, [0 1]); then output is [ 0 0 1 1] according to attached table.

[Jimmy cho]

Omg!

Sorry for this, you're right , I just messed it up.

thanks alot , didn't notice omg!

Im editing what I wrote:

You wrote in the question (I quote without changing a single word)

"binrestore(2,[0 0 1 1]) ----> output= [0 1 ]

"for example of the function that Im trying to implement, If I input to it:

...

reversebinrestore(2, [0 1 ]) --> Output=[0 0 1 1]"

Now if you said "you didn't understand me" then "you didn't understand you" since we just did exactly what you request in this question:

"reversebinrestore(2, [0 1 ]) --> Output=[0 0 1 1]"

""reversebinrestore(4, [0 1 ]) --> Output=[1 1 0 0 0 0 1 1]"

the output is accordingly to the table ..

Now is it understandable? your code is works fine but not the required output, my bad sorry!

So my reversebinrestore does as what I mentioned here above (the reverse ..)

[Jimmy cho]

should I edit again my thread or it's understandable? sorry for my missunderstandings.

Answer 2

[Bruno Luong]

function [output, outputchar] = reversebinrestore(m, array)

if ischar(array)
    array = array-'0';
end

switch m
    case 1
        map0 = [0];
    case 2
        map0 = [0 0];
    case 4
        map0 = [1 1 0 0];
    case 8
        map0 = [1 1 0 0 1 1 0 0];
    otherwise
        error('Multiplier not 1, 2, 4, 8')
end
map1 = 1-map0;
map = [map0; map1];

array = reshape(array, size(map,2), []).';
[b,output] = ismember(array, map, 'rows');
if ~all(b)
    error('invalid array');
end
output = output(:)'-1;
outputchar = char(output+'0');

end

[Bruno Luong]

You wrote in the question (I quote without changing a single word)

"binrestore(2,[0 1]) ----> output= [0 0 1 1 ]"

"for example of the function that Im trying to implement, If I input to it:

...

reversebinrestore(2, [0 0 1 1 ]) --> Output=[0 1]"

Now if you said "you didn't understand me" then "you didn't understand you" since we just did exactly what you request in this question:

"reversebinrestore(2, [0 0 1 1 ]) --> Output=[0 1]"

 >> reversebinrestore(2, [0 0 1 1 ]) 

ans =

     0     1

[Jimmy cho]

Hi once again, my bad!

Omg, Sorry for this, you're right , I just messed it up.

thanks alot , didn't notice omg!

Im editing what I wrote:

You wrote in the question (I quote without changing a single word)

"binrestore(2,[0 0 1 1]) ----> output= [0 1 ]

"for example of the function that Im trying to implement, If I input to it:

...

reversebinrestore(2, [0 1 ]) --> Output=[0 0 1 1]"

Now if you said "you didn't understand me" then "you didn't understand you" since we just did exactly what you request in this question:

"reversebinrestore(2, [0 1 ]) --> Output=[0 0 1 1]"

""reversebinrestore(4, [0 1 ]) --> Output=[1 1 0 0 0 0 1 1]"

the output is accordingly to the table ..

Now is it understandable? your code is works fine but not the required output, my bad sorry!

So my reversebinrestore does as what I mentioned here above (the reverse ..)

[Jimmy cho]

should I edit again my thread or it's understandable? sorry for my missunderstandings.

[Bruno Luong]

I let you sortout the filename xxx, yyy you seem to describe them one way or another depending on your mood.

function map = getmap(m)
switch m
    case 1
        map0 = [0];
    case 2
        map0 = [0 0];
    case 4
        map0 = [1 1 0 0];
    case 8
        map0 = [1 1 0 0 1 1 0 0];
    otherwise
        error('Multiplier not 1, 2, 4, 8')
end
map1 = 1-map0;
map = [map0; map1];
end

Then xxx

function [output, outputchar] = xxx(m, array)
if ischar(array)
    array = array-'0';
end
map = getmap(m);
output = map(array+1,:);
output = reshape(output',1,[]);
outputchar = char(output+'0');
end

Then yyy

function [output, outputchar] = yyy(m, array)
if ischar(array)
    array = array-'0';
end
map = getmap(m);
array = reshape(array, size(map,2), []).';
[b,output] = ismember(array, map, 'rows');
if ~all(b)
    error('invalid array');
end
output = output(:)'-1;
outputchar = char(output+'0');
end

These is examples of what xxx and yyy do, one is the reverse of the other. Rename it as your mood tells.

>> yyy(2,[0 0 1 1])

ans =

     0     1

>> xxx(2,[0 1])

ans =

     0     0     1     1

>> 

[Jimmy cho]

100% !

but for function xxx, why do I need two outputs ? I mean

function [output, outputchar] = xxx(m, array)

the variable called output is the array output of my function, but what about the second output called outputchar .. do I need it?!

because "ans" is actually the output array of function xxx that's stored on variable array called "output" ..am I right?!

thanks alot for any clarification on the code

[Bruno Luong]

If you need only once just call with once.

In YOUR code you returns a char array

ret=num2str(array,'%d');

Then in your description youu said

"output= [0 1 ]"

So I don't know which format you decide to use so I make both available.

Decide then stick on what you have decided. This appply on data format, function name, filename, etc...

You seem to hesitate a lot.

[Jimmy cho]

I understand you!

but in the case function [output, outputchar] = xxx(m, array)

it returns just an array, doesn't return two @outputs!

So maybe Im missing something? I mean if I write xxx(1 , [0 1])

it must return in your case two outputs one for output , the other for outputcar

but once I write on command window xxx(1 , [0 1]) it just returns one output and it's an array of integers and not two @outputs although it should return output , outputchar.

could you explain why?!

thanks alot for your help

[Bruno Luong]

Depend how you call it. This is basic MATLAB calling syntax and mmore advanced ignore output

>> xxx(2,[0 1])

ans =

     0     0     1     1

>> out = xxx(2,[0 1])

out =

     0     0     1     1

>> [out, outstr] = xxx(2,[0 1])

out =

     0     0     1     1


outstr =

    '0011'

>> [~, outstr] = xxx(2,[0 1])

outstr =

    '0011'

>> [~, outstr] = xxx(2,'01')

outstr =

    '0011'

>>