blob: ba1a42f41ece2d4f7d34e2d0b74dfb384901caeb [file] [log] [blame]
/*
* Copyright (C) 2004-2010 NXP Software
* Copyright (C) 2010 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "BIQUAD.h"
#include "BP_1I_D16F32Cll_TRC_WRA_01_Private.h"
#include "LVM_Macros.h"
/**************************************************************************
ASSUMPTIONS:
COEFS-
pBiquadState->coefs[0] is A0,
pBiquadState->coefs[1] is -B2,
pBiquadState->coefs[2] is -B1, these are in Q30 format
DELAYS-
pBiquadState->pDelays[0] is x(n-1)L in Q0 format
pBiquadState->pDelays[1] is x(n-2)L in Q0 format
pBiquadState->pDelays[2] is y(n-1)L in Q16 format
pBiquadState->pDelays[3] is y(n-2)L in Q16 format
***************************************************************************/
void BP_1I_D16F32C30_TRC_WRA_01 ( Biquad_Instance_t *pInstance,
LVM_INT16 *pDataIn,
LVM_INT16 *pDataOut,
LVM_INT16 NrSamples)
{
LVM_INT32 ynL,templ;
LVM_INT16 ii;
PFilter_State pBiquadState = (PFilter_State) pInstance;
for (ii = NrSamples; ii != 0; ii--)
{
/**************************************************************************
PROCESSING OF THE LEFT CHANNEL
***************************************************************************/
// ynL= (A0 (Q30) * (x(n)L (Q0) - x(n-2)L (Q0) ) >>14) in Q16
templ= (LVM_INT32) *pDataIn-pBiquadState->pDelays[1];
MUL32x32INTO32(pBiquadState->coefs[0],templ,ynL,14)
// ynL+= ((-B2 (Q30) * y(n-2)L (Q16) ) >>30) in Q16
MUL32x32INTO32(pBiquadState->coefs[1],pBiquadState->pDelays[3],templ,30)
ynL+=templ;
// ynL+= ((-B1 (Q30) * y(n-1)L (Q16) ) >>30) in Q16
MUL32x32INTO32(pBiquadState->coefs[2],pBiquadState->pDelays[2],templ,30)
ynL+=templ;
/**************************************************************************
UPDATING THE DELAYS
***************************************************************************/
pBiquadState->pDelays[3]=pBiquadState->pDelays[2]; // y(n-2)L=y(n-1)L
pBiquadState->pDelays[1]=pBiquadState->pDelays[0]; // x(n-2)L=x(n-1)L
pBiquadState->pDelays[2]=ynL; // Update y(n-1)L in Q16
pBiquadState->pDelays[0]=(*pDataIn++); // Update x(n-1)L in Q0
/**************************************************************************
WRITING THE OUTPUT
***************************************************************************/
*pDataOut++=(LVM_INT16)(ynL>>16); // Write Left output in Q0
}
}