vec_math
Class MaximumEntropy

java.lang.Object
  vec_math.LinearPrediction
      vec_math.MaximumEntropy

public class MaximumEntropy
extends LinearPrediction

A class that uses maximum entropy to estimate a power spectrum. It follows the description in Num. Rec. 13.7, p572ff..

Nested Class Summary

Nested classes/interfaces inherited from class vec_math.LinearPrediction

LinearPrediction.Debug

Field Summary

private double

sampling The sampling interval.

Constructor Summary

MaximumEntropy()
Constructs a new maximum entropy object with default bias removing as in the parental class.

MaximumEntropy(boolean subtract)
Constructs a maximum entropy object.

Method Summary

double	power(double f) Returns the power spectrum estimate at the given frequency.
void	setSampling(double dt) Sets the sampling intervall.

Methods inherited from class vec_math.LinearPrediction

calculateCoefficients, getCoefficients, getData, getDiscrepancy, predict, setData

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

sampling

private double sampling

The sampling interval.

Constructor Detail

MaximumEntropy

public MaximumEntropy()

Constructs a new maximum entropy object with default bias removing as in the parental class.

MaximumEntropy

public MaximumEntropy(boolean subtract)

Constructs a maximum entropy object. If the supplied boolean is true, the data mean is subtracted prior to the calculation of the LP coefficients and added again on the predicted values.

Parameters:

subtract - Whether to subtract the mean.

Method Detail

setSampling

public void setSampling(double dt)

Sets the sampling intervall.

Parameters:

dt - Sampling intervall in seconds.

power

public double power(double f)

Returns the power spectrum estimate at the given frequency. Coefficients must have been calculated before and the sampling interval must have been set. Frequencies should always lie below the Nyquist frequency, though this is not checked.

Parameters:

f - Frequency where the power should be estimated in Hz.