net.java.sip.communicator.impl.neomedia.codec.audio.silk
Class Silk_residual_energy_FLP

java.lang.Object
  net.java.sip.communicator.impl.neomedia.codec.audio.silk.Silk_residual_energy_FLP

public class Silk_residual_energy_FLP
extends Object

Residual energy.

Author:

Dingxin Xu

Field Summary

(package private) static int	MAX_ITERATIONS_RESIDUAL_NRG
(package private) static float	REGULARIZATION_FACTOR

Constructor Summary

Silk_residual_energy_FLP()

Method Summary

(package private) static float	SKP_Silk_residual_energy_covar_FLP(float[] c, int c_offset, float[] wXX, int wXX_offset, float[] wXx, float wxx, int D) Residual energy: nrg = wxx - 2 * wXx * c + c' * wXX * c.
(package private) static void	SKP_Silk_residual_energy_FLP(float[] nrgs, float[] x, float[][] a, float[] gains, int subfr_length, int LPC_order) Calculates residual energies of input subframes where all subframes have LPC_order of preceeding samples

Methods inherited from class java.lang.Object

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

Field Detail

MAX_ITERATIONS_RESIDUAL_NRG

static final int MAX_ITERATIONS_RESIDUAL_NRG

See Also:

Constant Field Values

REGULARIZATION_FACTOR

static final float REGULARIZATION_FACTOR

See Also:

Constant Field Values

Constructor Detail

Silk_residual_energy_FLP

public Silk_residual_energy_FLP()

Method Detail

SKP_Silk_residual_energy_covar_FLP

static float SKP_Silk_residual_energy_covar_FLP(float[] c,
                                                int c_offset,
                                                float[] wXX,
                                                int wXX_offset,
                                                float[] wXx,
                                                float wxx,
                                                int D)

Residual energy: nrg = wxx - 2 * wXx * c + c' * wXX * c.

Parameters:

c - Filter coefficients

c_offset - offset of valid data.

wXX - Weighted correlation matrix, reg. out

wXX_offset - offset of valid data.

wXx - Weighted correlation vector

wxx - Weighted correlation value

D - Dimension

Returns:

Weighted residual energy

SKP_Silk_residual_energy_FLP

static void SKP_Silk_residual_energy_FLP(float[] nrgs,
                                         float[] x,
                                         float[][] a,
                                         float[] gains,
                                         int subfr_length,
                                         int LPC_order)

Calculates residual energies of input subframes where all subframes have LPC_order of preceeding samples

Parameters:

nrgs - Residual energy per subframe

x - Input signal

a - AR coefs for each frame half

gains - Quantization gains

subfr_length - Subframe length

LPC_order - LPC order