<< dirname 6-1-1 ric_desc >>
removed >> removed > 6-1-1 > noisegen

noisegen

noise generation

Syntax

b = noisegen(pas, Tmax, sig)

Arguments

pas

real scalar, the time increment

Tmax

real scalar, the final time of the interval of noise generation

sig

real scalar, the standard deviation of the noise

Description

noisegen() is removed Examples (clearer, shorter, and more powerful) using grand() follow:

100 10-point wide steps with random amplitudes of normal distribution of mean 1 and standard deviation 0.7:

u = grand(1,100,"nor",1,0.7) .*. ones(1,10);
clf, plot(u);

40 15-point wide steps with random amplitudes of uniform distribution in [1 3], with initial level 0.5 and final level hold:

width = 15;
u = grand(1,40, "unf", 1, 3) .*. ones(1,width);
u = [ones(1,8*width)*0.5, u, ones(1,5*width)*u($)];
clf, plot(u);

noisegen() generates a Scilab function b = Noise(t) where Noise(t) is a piecewise constant function (constant on [k*pas, (k+1)*pas]). The value on each constant interval are random values from i.i.d Gaussian variables of standard deviation sig. The function is constant for t ≤ 0 and t ≥ Tmax.

grand allows to generate noises according to a wide variety of probability distributions.

Examples

Example #1: noisegen()

noisegen(0.5, 30, 1.0);
x = -5:0.01:35;
y = feval(x, Noise);
plot(x, y);

Example #2: rand()

// Plot a zero mean gaussian white noise with the variance 1.
// To use a different variance, multiply rand() by the square root of the variance.
t = 0:.5:30;
noise = grand(t, "nor", 0,1);
plot2d2(t, noise);

See also

History

VersionDescription
6.1.0 noisegen() is obsolete. It can be replaced using grand and .*..
6.1.1 noisegen() will be removed in the next Scilab version.

<< dirname 6-1-1 ric_desc >>