hello list ,
i am pranay ,undergraduate student at Indian Institute of Technology ,Madras
.i am wondering whether anyone in this list has worked on OBSns
(http://wine.icu.ac.kr/~obsns/index.php
<http://wine.icu.ac.kr/%7Eobsns/index.php>) software for NS .
I am studying the way in which self similar characteristic is implemented in
obsns .
But i cant understand ,how the flicker noise is generated in the file
fract_flick.cc
int N; /* filter order */
int i;
double f, ae, be;
double *apt, *bpt;
double W, R;
N = (int)ceil (n*h);
*a = 0.13; /* From Barnes and Greenhall paper */
R = pow (10.0, 1/(2.0*h));
W = ((1.0 - (*a))/sqrt(*a));
*b = 0.0;
for (i = 2, apt = a+1, bpt = b+1; i <= N; i++, apt++, bpt++)
{
W = W / pow (R, 2.0 - gamma);
(*bpt) = 1.0 + ((0.5*W) * (W - sqrt((W * W) + 4.0)));
W = W / pow (R, gamma);
(*apt) = 1.0 + ((0.5*W) * (W - sqrt((W * W) + 4.0)));
}
}
void flicker (int N, double *y, double *y_old, double *a, double *b)
/*
* Inegrate fractional flicker noise generator over 1 time step
*
* N = number of filter stages (= n*h, see init_flick above)
* y = pointer to array of current state vector plus white noise input in
* element zero (array size = N+1)
* y_old = pointer to array of previous time step state vector plus white
* noise input in element zero (array size = N+1)
* a = pointer to array of poles (array size = N)
* (phi array of Barnes and Greenhall)
* b = pointer to array of zeros (array size = N)
* (theta array of Barnes and Greenhall)
*/
{
int i;
double *ypt, *y_oldpt, *apt, *bpt;
*(y+1) = ( (*a) - (*b) ) * (*(y_old+1)) + (*(y));
for (i = 2, ypt = y+2, y_oldpt = y_old+2, apt = a+1, bpt = b+1; i <=
N;
i++, ypt++, y_oldpt++, apt++, bpt++)
{
*ypt = (*apt) * (*y_oldpt) + ( (*(ypt-1)) -
(*bpt) * (*(y_oldpt - 1)) );
}
for (i = 0, ypt = y, y_oldpt = y_old, apt = a, bpt = b; i <= N;
i++, ypt++, y_oldpt++, apt++, bpt++)
*y_oldpt = (*ypt);
}
its referring to some paper ,(barnes and greenhall) ,but i am not able to
find that paper on net or any library .
if someone can please explain me
W = W / pow (R, 2.0 - gamma);
(*bpt) = 1.0 + ((0.5*W) * (W - sqrt((W * W) + 4.0)));
W = W / pow (R, gamma);
(*apt) = 1.0 + ((0.5*W) * (W - sqrt((W * W) + 4.0)));
why these formulae are used for generating arrrrays 'a' and 'b', i ud be
very thankful .
or else if u can send me the pdf of the above mentioned paper ,that ud be
great .
please help me ,as this is crucial for my project .
thank you
Pranay Prateek
Department of Electrical Enginnering
Indian Institute of Technology ,Madras
Chennai ,India .
.
