# Inserting new element after each element of an array

9 views (last 30 days)
Neeraj Chimwal on 25 Mar 2021
Edited: DGM on 26 Mar 2021
I have an array arr = [2,4,6]
After converting this array elements to binary using de2bi(arr,'left-msb'), I get
0 1 0
1 0 0
1 1 0
Now what I want to do is to insert 0 after each 0 and 1 after each 1. So the result would be
0 0 1 1 0 0
1 1 0 0 0 0
1 1 1 1 0 0
I tried looping through the array, but since length of binary array is still 3, I can't loop through each element.

Stephen23 on 25 Mar 2021
The MATLAB approach:
arr = [2,4,6];
mat = repelem(de2bi(arr,'left-msb'),1,2)
mat = 3×6
0 0 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0

David Goodmanson on 25 Mar 2021
Edited: David Goodmanson on 25 Mar 2021
Hi Neeraj,
not having the communications toolbox I used dec2bin instead, which gives a character array but it is basically the same idea and should work on a binary array I would think.
arr = [2,4,6];
a = dec2bin(arr)
s1 = size(a,1);
s2 = size(a,2);
c(1:s1,1:2:2*s2) = a;
c(1:s1,2:2:2*s2) = a
Neeraj Chimwal on 25 Mar 2021
that worked. If you don't mind can you please explain me the last two steps
David Goodmanson on 26 Mar 2021
Hi Neeraj,
The index 1 : 2*s2 is twice as large as the number of columns in 'a'.
In the first step, by using 1:2:2*s2, 'a' is inserted into the odd numbered columns of the new matrix.
In the 2nd step, by using 2:2:2*s2, 'a' is inserted into the even numbered columns of the new matrix.

DGM on 25 Mar 2021
You should be able to loop through the array regardless of its size. That said, you don't need to.
A=[1 2 3 4];
B=dec2bin(A);
expanded=B(:,round(0.5:0.5:size(B,2)));
which gives:
expanded =
000011
001100
001111
110000
Neeraj Chimwal on 25 Mar 2021
that worked. Can you please explain the last step? I didn't understand what 0.5:0.5 did here
DGM on 26 Mar 2021
Edited: DGM on 26 Mar 2021
This bit:
0.5:0.5:size(B,2)
generates a vector like so:
[0.5 1.0 1.5 2.0 2.5 3.0 ... ]
rounding that vector gives us this:
[1 1 2 2 3 3 ... ]
so that we're referencing each element of B twice
Alternatively, you could use something like kron() to generate the same index vector:
kron([1 2 3 4],[1 1])
would yield
[1 1 2 2 3 3 4 4 ]