Struct light_curve_feature::periodogram::Periodogram

pub struct Periodogram<T> { /* fields omitted */ }

Lamb-Scargle periodogram calculator on uniform frequency grid

Frequencies are given by $\omega = {\min\omega..\max\omega}$: $N_\omega$ nodes with step $\Delta\omega$: $\min\omega = \Delta\omega$, $\max\omega = N_\omega \Delta\omega$.

Parameters of the grid can be derived from time series properties: typical time interval $\delta t$ and duration of observation. The idea is to set maximum frequency to Nyquist value $\pi / \Delta t$ and minimum frequency to $2\pi / \mathrm{duration}$, while nyquist and resolution factors are used to widen this interval: $$ \max\omega = N_\omega \Delta\omega = \frac{\pi}{\Delta t}, $$ $$ \min\omega = \Delta\omega = \frac{2\pi}{\mathrm{resolution} \times \mathrm{duration}}. $$

Implementations

impl<T> Periodogram<T> where
    T: Float, 

pub fn new(
    periodogram_power: Box<dyn PeriodogramPower<T>>,
    freq_grid: FreqGrid<T>
) -> Self

pub fn set_periodogram_power(
    &mut self,
    periodogram_power: Box<dyn PeriodogramPower<T>>
) -> &mut Self

pub fn init_thread_local_fft_plans(n: &[usize])

pub fn from_t(
    periodogram_power: Box<dyn PeriodogramPower<T>>,
    t: &[T],
    resolution: f32,
    max_freq_factor: f32,
    nyquist: &Box<dyn NyquistFreq<T>>
) -> Self

pub fn freq(&self, i: usize) -> T

pub fn power(&self, ts: &mut TimeSeries<T>) -> Vec<T>