nafflib#

Functions

`calcnaff()`	Computes NAFF decomposition for a phase space trajectory
`naff_cc()`	Compile nafflibrary for Matlab
`naff_example()`	Example to test naff within matlab

calcnaff(y, yp, win)#: Computes NAFF decomposition for a phase space trajectory

[nu amp phase] = calcnaff(y, yp, win)

INPUTS

1. Y - position vector

2. Yp - angle vector

3. WindowType - Window type - 0 {Default} no windowing

1 Window of Hann

2 etc

4. nfreq - Maximum number of fundamental frequencies to search for

10 {Default}

5. debug - 1 means debug flag turned on

0 {Default}

Optional Flags

‘Debug’ - turn on deubbing flag

‘Display’ - print ou results

‘Hanning’ - select Window of Hann, WindowType = 1

‘Raw’ or ‘NoWindow’ - select Window of Hann, WindowType = 0

OUTPUTS

1. frequency - frequency vector with sorted amplitudes

by default the algorithm try to compute the 10 first fundamental

frequencies of the system.

2. amplitude - amplitudes associated with fundamental frequencies

3. phase - phases associated with fundamental frequencies

NOTES

1. Mimimum number of turns is 64 (66)

2. Number of turn has to be a multiple of 6 for internal optimization

reason and just above a power of 2. Example 1026 is divived by 6 and

above 1024 = pow2(10)

Examples

NT = 9996; % divisible by 6

simple quasiperiodic (even period) motion

y =2+0.1*cos(pi*(0:NT-1))+0.00125*cos(pi/3*(0:NT-1));

yp=2+0.1*sin(pi*(0:NT-1))+0.00125*sin(pi/3*(0:NT-1));

[nu ampl phase] = calcnaff(y,yp,1); % with windowing