Correct low-bit sampled power measurements and correlations. More...

#include <VanVleck.h>

Public Member Functions
	VanVleck ()
	Default constructor.
void	set_nlevel (int n)
	Set number of levels.
int	get_nlevel () const
	Get number of levels;.
void	set_uniform_threshold ()
	Set uniform sampling thresholds.
void	set_threshold (float *thr)
	Set arbitrary sampling thresholds.
void	set_uniform_levels ()
	Set uniform output levels.
void	set_canonical_levels ()
	Set "canonical" output levels.
void	set_levels (float *lev)
	Set arbitrary output levels.
double	out_pow (double in_pow, double in_dc=0.0)
	Return output (quantized) power given input power, dc.
double	d_out_pow (double in_pow, double in_dc=0.0)
	Return derivative d(out_pow)/d(in_pow)
double	in_pow (double out_pow)
	Return input (analog) power given output (quantized) power.
void	set_in_power (double px, double py)
	Set input powers for correlation correction.
void	correct_acf (std::vector< float > data)
	Correct a vector of auto-correlations using current pow_x.
void	correct_ccf (std::vector< float > data)
	Correct a vector of cross-correlations using current input powers.

Protected Member Functions
void	compute_coeffs ()
	Compute coefficients for correlation correction using pow_x, pow_y.

Protected Attributes
int	nlevel
	Number of levels.
double	pow_x
	Input powers to use in correcting corrrelations.
double	pow_y
std::vector< float >	thresh
	Vector of sampling thresholds.
std::vector< float >	levels
	Vector of output levels.
double	c1x
	Storage for coeffs we may want to reuse.
double	c1y
double	c3x
double	c3y

Detailed Description

Correct low-bit sampled power measurements and correlations.

This class contains routines to rescale power and correlation measurements made from N-level sampled data. The number of levels, and their output values are arbitrary (but should be constant). The only constraint is that both thresholds and output levels must be symmetric about 0. Quantized (pre-correction) data should be scaled so that it has power levels between $L_0^2$ and $L_{max}^2$ , where $L_0$ ( $L_{max}$ ) is the smallest (largest) level in absolute value. Input power values are returned in the same units as the threshold values.