How to create a binary matrix

snake eyes (view profile)

on 20 Jul 2011

I want to make matrix whose each element's size will be 1 bit, i.e, it will be a binary matrix.How can I do it?

Products

No products are associated with this question.

Jan Simon (view profile)

on 20 Jul 2011

There is not BIT type in Matlab. Usually binary arrays are stored as LOGICAL, which are equivalent to UINT8. E.g.:

```B = rand(16, 16) > 0.8;
whos('B')
```

Now you can pack this in blocks of 8 elements to a UINT8 array:

```T = 2 .^ (0:7).';
B8 = transpose(reshape(B, 8, []));
BitArray = uint8(B8 * T);
```

Friedrich (view profile)

on 20 Jul 2011

Hi,

this is not possible in MATLAB and in other languages too like C or C++. In modern computer architectures, a byte is the smallest addressable unit of memory. To pack multiple bits into a byte requires applying extra bit-shift operations. At the compiler level, it's a trade off of memory vs. speed requirements (and in high-performance software, those extra bit-shift operations can add up and slow down the application needlessly).

snake eyes

snake eyes (view profile)

on 20 Jul 2011

C has a feature known as 'bit field' where I can specify custom length.

Walter Roberson

Walter Roberson (view profile)

on 20 Jul 2011

C does not specify whether bit-fields start from the beginning or the end of a word, and C permits but does not require bit-fields to cross words. In C, arrays of bit-fields are arrays of words: the "packed" attribute possibility was removed from the C language before C was standardized. There is thus no binary matrix in C, just arrays of structs whose members are bit fields.

Walter Roberson

Walter Roberson (view profile)

on 20 Jul 2011

Well, in _popular_ computer architectures; bit-sliced and bit-addressable architectures still have their purposes.

Sean de Wolski (view profile)

on 20 Jul 2011

To get logical binary values either use:

```x = logical(x)
```

or any conditional operator e.g.

```x = x>7
x = x==11;
```