CODEWITHSUNDEEP
arraysperformancematlabloopsoctave
ArthurChamz
ArthurChamz

Reputation: 2049

Fast loop alternative for array gen and concat in MATLAB / Octave

I have a large file with values, I'm trying to create a waveform with those values.

It takes too long, and since I'm not the best at Octave, I think there may be another way to do this. I'm looking for a faster alternative, one that uses Octave's array handling capabilities instead of me doing a loop, which I assume is slow.

EDIT

csv is a [1xn] array.

csv = csvread('values.csv');

n = length(csv);

t = [0 : 1 / (freq * spc) : 1 / freq];
t = t(2 : length(t));

wave = csv(1) * sin(two_pi_freq * t);

for i = 2 : n

    wave = [wave (csv(i) * sin(two_pi_freq * t))];

endfor

Upvotes: 1

Views: 359

Answers (1)

Hoki
Hoki

Reputation: 11812

Edited answer based on latest comments:

To avoid any for loop, you can multiply your two vectors to obtain an array, then reshape the array to put all the values in line.

The way you are building t, it will be of dimension [1xm] (with m=length(t))

For a csv vector of dimension [1xn], instead of your loop, use:

%// generate an array size [nxm] by multiplying [nx1] * [1xm]
wave2 = csv.' * sin(two_pi_freq * t) ;    
wave2 = reshape( wave2.' , [] , length(t)*length(csv) ) ; %// reshape the array in one [1xm*n] line

This should replace your last 4 lines of code (the first definition of wave and the 3 lines of the loop.

Upvotes: 2

Related Questions