out of memory

hello friends, i have a problem with a code section for receiver synchronization. When i run the code, an error is occured. That error is :

??? Error using ==> mtimes

Out of memory. Type HELP MEMORY for your options.

Error in ==> alici at 21

r1 = r*cos(2*pi*f_isaret*t);

load('kayit_alici.mat','r','veri')

f_isaret = 40000;

T_isaret = 1 / f_isaret;

f_AI_sample = 160000;

T_AI_sample = 1/f_AI_sample;

T_bit = 2*power(10,-3);

T_frame = 1;

bosluk = 20;

barker_uz = 13;

veri_uz = T_frame/T_bit - 1 - bosluk - barker_uz;

Hiz = 1 / T_bit;

sample_number_per_bit = T_bit * f_AI_sample;

h = [1 1 1 1 1 -1 -1 1 1 -1 1 -1 1];

h_barker = fliplr(h);

N = sample_number_per_bit;

h_barker2 = reshape(repmat(h_barker,N,1),1,N*length(h_barker));

t = 0:T_AI_sample:T_frame-T_AI_sample;

r1 = r*cos(2*pi*f_isaret*t);

r2 = -j*r*sin(2*pi*f_isaret*t);

rp = r1 + r2;

rp_conv = conv(rp,h_barker2);

rp_c_n = norm(rp_conv);

max_rp_c_n = max(rp_c_n);

% dimension of r is 160000x1

waiting for your helps :)

Answer 1

[Walter Roberson]

Probably, Change

f_isaret*t

to

f_isaret.*t

[osman yurdakul]

i tried it a bit ago but the error was the same :(

[Walter Roberson]

Sorry I didn't look closely enough, f_isaret is not a problem.

However, you load "r" from a file and so we cannot get any hint of what size it is. If it is a vector or array, then you should probably be using r.* instead of r* when calculating r1 and r2.

[osman yurdakul]

yes, i load 'r' from a mat-file the size of it is 160000x1. 'r' array is samples that have been gotten from a signal that its time 1 sec and the samplig freq is 160 Khz

[Matt Tearle]

Yup, Walter's right - you need an array multiply there (.*). r is a column vector, but t is a row vector, so cos(...*t) is also a row vector. * is matrix multiply in MATLAB, so it's trying to make r1 a matrix of size 160000-by-length(t). Solution, transpose either t or r and do an array multiply

r1 = r.*cos(...*t');

Same on the next line.

[Walter Roberson]

Your r is a column vector and your t is a row vector and you are using * which is matrix multiplication. The result would be an array that was 160000 by length(t), which you probably don't want.

But for r.*t to work, r and t would have to have the same number of elements, which it does not appear that your program would promise: for subtle reasons, your t could end up one entry too short. You would be better of defining t as

t = linspace(0,T_frame,f_AI_sample+1);

t(end) = [];

This would have the correct length for element-by-element multiplication by a vector of length f_AI_sample .

[Walter Roberson]

If you transposed t but used * then you would get an error about inner dimensions must agree. If you transposed r but used * then you would get a single value as output, not a vector the same length as t.

[osman yurdakul]

thank you friends it worked :) really thank you... i get it now :)