how to create loops to automatically select next bits??

a = [1 0 1 0 1 1 1 1 1 0 0 0 0 0......] <--- 1 row 80 columns

I am trying to create a loop such that mat b has 1st 16 bits of mat a in first run and in next run the next 16 bits of mat a

please help me in how to write a code for such loop

should i use for loop or while loop

Kindly help

[Guillaume]

Most likely, you shouldn't be using a loop at all. You probably can do whatever it is you want to do on all 16-bit blocks all at once. What is it you want to do?

[Ayesha Punjabi]

a = [1 0 1 0 1 0 0 1 0 1 0.... so mat a is 1 row and 80 columns of binary bits

in mat b i want to put 1st 16 bits of data from mat a and perform a mutiplication with another matrix which is 16 bits of binary data example c [1 1 1 1 0 0 0 0 1 0 1 0 0 0 0 0]

this was in first run

in next run I want to again put next 16 bits from mat a to mat b. use mat b to multiply with mat c

continue this process untill all the 80 bits are performed with multiplication with mat c but 16 at one time

[Jan]

@Ayesha: By teh way, you can omit the "mat", because nobody else use this expression. Usually "mat" means a ".mat" file.

Answer 1

[Jan]

a = randi([0,1], 1, 80);
for k = 1:16:80
    b = a(k:k+15);
    ...
end

Or:

aa = reshape(a, 16, 5);
for k = 1:5
    b = aa(:, k).';
    ...
end

Or:

for k = 1:5
    b = a((1 + (k-1) * 16):(k * 16));
    ...
end

[Guillaume]

Or, as I commented on the question, probably don't use a loop at all.

[Ayesha Punjabi]

thank you for your suggestions.

could you please let me know the . . . in the code. I tried running the code but it is showing just 80 as the output

[Jan]

@Ayesha: My code should not show anything as output. The "..." is the correct location to insert, what you want to be computed. You did not mention, what you want to happen with the 16 bit blocks of the vector. I've showed you some methods to obatin the 16 bit block in the variable b, but now it is your turn to insert the calculations with b.

Now you have mentioned, that you want to multiply b with the vector c - an elementwise or matrix multiplication?

c      = [1 1 1 1 0 0 0 0 1 0 1 0 0 0 0 0];
aa     = reshape(a, 16, 5);
result = zeros(1, 5);
for k = 1:5
    b = aa(:, k);
    result(k) = c * b;  % Matrix multiplication ==> scalar output
end

But as said by Guillaume already: This works much nicer without a loop:

result = c * reshape(a, 16, 5)

Answer 2

[KALYAN ACHARJYA]

x=randi([0 1],1,80); %Here I choosed random, consider your actual a
c=[1 1 1 1 0 0 0 0 1 0 1 0 0 0 0 0]; %As given by you
for i=1:16:length(x)-15;
    mat_b=x(i:i+15).*c;
    fprintf('\n The start bit position: %d, mat b result:',i);
    disp(mat_b);
end

Answer 3

[Guillaume]

in mat b i want to put 1st 16 bits of data from mat a and perform a mutiplication with another matrix which is 16 bits of binary in next run I want to again put next 16 bits from mat a to mat b. use mat b to multiply with mat c continue this process untill all the 80 bits

As I expected for this you don't need a loop. Simply reshape your a in columns of 16 bit. And multiply at once all the columns with the same c:

a = randi([0 1], 1, 80);  %row vector of 80 bits. demo data
c = [1 1 1 1 0 0 0 0 1 0 1 0 0 0 0 0];  %row vector

result = reshape(a, 16, []) .* c(:);

Each column of result correspond to a group of 16 bit. If you want it back as a 1x80 vector:

reshape(result, 1, [])