DFT (discrete fourier transform) with C++ STL

c++ stl dft

13,884

std::cout << abs(sum)/N << std::endl;

Why are you dividing by N?

The coefficients are without the division. See the wiki.

These are the values I get with Matlab:

12.0000000000000 + 0.00000000000000i
-0.971586454726535 - 0.153884176858763i
-4.26246117974981 - 1.38495759172886i   
-0.0712959999079796 - 0.0363271264002681i   
-0.473606797749979 - 0.344095480117793i
0.00000000000000 + 0.00000000000000i
-0.237538820250189 - 0.326944137602412i
0.0185095954079375 + 0.0363271264002681i
-0.0263932022500213 - 0.0812299240582274i
0.0243728592265771 + 0.153884176858763i
0.00000000000000 + 0.00000000000000i    
0.0243728592265771 - 0.153884176858763i
-0.0263932022500213 + 0.0812299240582274i
0.0185095954079375 - 0.0363271264002681i    
-0.237538820250189 + 0.326944137602412i 
0.00000000000000 + 0.00000000000000i    
-0.473606797749979 + 0.344095480117793i 
-0.0712959999079796 + 0.0363271264002681i
-4.26246117974981 + 1.38495759172886i   
-0.971586454726535 + 0.153884176858763i

and I see more or less the same values when I print using std::cout << sum << std::endl;

13,884

Author by

Johannes

Updated on July 25, 2022

Comments

Johannes almost 2 years

I try to calculate the DFT for this array x_1. It must be dead simple, but my values are way too low. What's wrong with my code?

Please no links to other examples - just looking for a fix for my own code.

#include <iostream>
#include <complex>
#include <cassert>

int main ()
{
    const unsigned int N = 20;

    const double x_1[N] = {0, 0.3, 0.6, 0.8, 1, 1, 0.9, 0.7, 0.5, 0.2, 0.2, 0.5, 0.7, 0.9, 1, 1, 0.8, 0.6, 0.3, 0};

    for(unsigned int k = 0; k < N; k++)
    {
        std::complex<double> sum(0.0,0.0);
        for(unsigned int j = 0; j < N; j++)
        {
            int integers = -2*j*k;
            std::complex<double> my_exponent(0.0, M_PI/N*(double)integers);
            sum += x_1[j] * std::exp(my_exponent);
        }
        std::cout << abs(sum)/N << std::endl;
    }
    return 0;
}

Johannes almost 12 years

I shall do this for my exercise. This can not be the reason of the error.
Johannes almost 12 years

Thanks for trying with matlab, then the calculation must be correct... Why is the first value (12.0) that high?
BertR almost 12 years

The DC value is the sum of all values. If you sum all values, you'll get 12.
BertR almost 12 years

@Johannes: Does this solve your problem? In case it does, can you maybe vote and accept the answer? Thanks :-)
Johannes almost 12 years

Actually, I don't know what's wrong with my results. But your MATLAB output shows that my code is correct. I'll find out more week, but I think you should get an "accept" :)
Johannes almost 12 years

Now I know. I made a copy and paste error getting the input. The algorithm, however, is completely correct. Thanks anyway.