dsp - CMSIS DSP Library from CMSIS 2.0. See http://www.…

Users » simon » Code » dsp

CMSIS DSP Library from CMSIS 2.0. See http://www.onarm.com/cmsis/ for full details

Dependents: K22F_DSP_Matrix_least_square BNO055-ELEC3810 1BNO055 ECE4180Project--Slave2 ... more

src/Cortex-M4-M3/FilteringFunctions/arm_fir_decimate_f32.c@0:1014af42efd9, 2011-03-10 (annotated)

Committer:: simon
Date:: Thu Mar 10 15:07:50 2011 +0000
Revision:: 0:1014af42efd9

Who changed what in which revision?

User	Revision	Line number	New contents of line
simon	0:1014af42efd9	1	/* ----------------------------------------------------------------------
simon	0:1014af42efd9	2	* Copyright (C) 2010 ARM Limited. All rights reserved.
simon	0:1014af42efd9	3	*
simon	0:1014af42efd9	4	* $Date: 29. November 2010
simon	0:1014af42efd9	5	* $Revision: V1.0.3
simon	0:1014af42efd9	6	*
simon	0:1014af42efd9	7	* Project: CMSIS DSP Library
simon	0:1014af42efd9	8	* Title: arm_fir_decimate_f32.c
simon	0:1014af42efd9	9	*
simon	0:1014af42efd9	10	* Description: FIR decimation for floating-point sequences.
simon	0:1014af42efd9	11	*
simon	0:1014af42efd9	12	* Target Processor: Cortex-M4/Cortex-M3
simon	0:1014af42efd9	13	*
simon	0:1014af42efd9	14	* Version 1.0.3 2010/11/29
simon	0:1014af42efd9	15	* Re-organized the CMSIS folders and updated documentation.
simon	0:1014af42efd9	16	*
simon	0:1014af42efd9	17	* Version 1.0.2 2010/11/11
simon	0:1014af42efd9	18	* Documentation updated.
simon	0:1014af42efd9	19	*
simon	0:1014af42efd9	20	* Version 1.0.1 2010/10/05
simon	0:1014af42efd9	21	* Production release and review comments incorporated.
simon	0:1014af42efd9	22	*
simon	0:1014af42efd9	23	* Version 1.0.0 2010/09/20
simon	0:1014af42efd9	24	* Production release and review comments incorporated
simon	0:1014af42efd9	25	*
simon	0:1014af42efd9	26	* Version 0.0.7 2010/06/10
simon	0:1014af42efd9	27	* Misra-C changes done
simon	0:1014af42efd9	28	*
simon	0:1014af42efd9	29	* -------------------------------------------------------------------- */
simon	0:1014af42efd9	30
simon	0:1014af42efd9	31	#include "arm_math.h"
simon	0:1014af42efd9	32
simon	0:1014af42efd9	33	/**
simon	0:1014af42efd9	34	* @ingroup groupFilters
simon	0:1014af42efd9	35	*/
simon	0:1014af42efd9	36
simon	0:1014af42efd9	37	/**
simon	0:1014af42efd9	38	* @defgroup FIR_decimate Finite Impulse Response (FIR) Decimator
simon	0:1014af42efd9	39	*
simon	0:1014af42efd9	40	* These functions combine an FIR filter together with a decimator.
simon	0:1014af42efd9	41	* They are used in multirate systems for reducing the sample rate of a signal without introducing aliasing distortion.
simon	0:1014af42efd9	42	* Conceptually, the functions are equivalent to the block diagram below:
simon	0:1014af42efd9	43	* \image html FIRDecimator.gif "Components included in the FIR Decimator functions"
simon	0:1014af42efd9	44	* When decimating by a factor of <code>M</code>, the signal should be prefiltered by a lowpass filter with a normalized
simon	0:1014af42efd9	45	* cutoff frequency of <code>1/M</code> in order to prevent aliasing distortion.
simon	0:1014af42efd9	46	* The user of the function is responsible for providing the filter coefficients.
simon	0:1014af42efd9	47	*
simon	0:1014af42efd9	48	* The FIR decimator functions provided in the CMSIS DSP Library combine the FIR filter and the decimator in an efficient manner.
simon	0:1014af42efd9	49	* Instead of calculating all of the FIR filter outputs and discarding <code>M-1</code> out of every <code>M</code>, only the
simon	0:1014af42efd9	50	* samples output by the decimator are computed.
simon	0:1014af42efd9	51	* The functions operate on blocks of input and output data.
simon	0:1014af42efd9	52	* <code>pSrc</code> points to an array of <code>blockSize</code> input values and
simon	0:1014af42efd9	53	* <code>pDst</code> points to an array of <code>blockSize/M</code> output values.
simon	0:1014af42efd9	54	* In order to have an integer number of output samples <code>blockSize</code>
simon	0:1014af42efd9	55	* must always be a multiple of the decimation factor <code>M</code>.
simon	0:1014af42efd9	56	*
simon	0:1014af42efd9	57	* The library provides separate functions for Q15, Q31 and floating-point data types.
simon	0:1014af42efd9	58	*
simon	0:1014af42efd9	59	* \par Algorithm:
simon	0:1014af42efd9	60	* The FIR portion of the algorithm uses the standard form filter:
simon	0:1014af42efd9	61	* <pre>
simon	0:1014af42efd9	62	* y[n] = b[0] * x[n] + b[1] * x[n-1] + b[2] * x[n-2] + ...+ b[numTaps-1] * x[n-numTaps+1]
simon	0:1014af42efd9	63	* </pre>
simon	0:1014af42efd9	64	* where, <code>b[n]</code> are the filter coefficients.
simon	0:1014af42efd9	65	* \par
simon	0:1014af42efd9	66	* The <code>pCoeffs</code> points to a coefficient array of size <code>numTaps</code>.
simon	0:1014af42efd9	67	* Coefficients are stored in time reversed order.
simon	0:1014af42efd9	68	* \par
simon	0:1014af42efd9	69	* <pre>
simon	0:1014af42efd9	70	* {b[numTaps-1], b[numTaps-2], b[N-2], ..., b[1], b[0]}
simon	0:1014af42efd9	71	* </pre>
simon	0:1014af42efd9	72	* \par
simon	0:1014af42efd9	73	* <code>pState</code> points to a state array of size <code>numTaps + blockSize - 1</code>.
simon	0:1014af42efd9	74	* Samples in the state buffer are stored in the order:
simon	0:1014af42efd9	75	* \par
simon	0:1014af42efd9	76	* <pre>
simon	0:1014af42efd9	77	* {x[n-numTaps+1], x[n-numTaps], x[n-numTaps-1], x[n-numTaps-2]....x[0], x[1], ..., x[blockSize-1]}
simon	0:1014af42efd9	78	* </pre>
simon	0:1014af42efd9	79	* The state variables are updated after each block of data is processed, the coefficients are untouched.
simon	0:1014af42efd9	80	*
simon	0:1014af42efd9	81	* \par Instance Structure
simon	0:1014af42efd9	82	* The coefficients and state variables for a filter are stored together in an instance data structure.
simon	0:1014af42efd9	83	* A separate instance structure must be defined for each filter.
simon	0:1014af42efd9	84	* Coefficient arrays may be shared among several instances while state variable array should be allocated separately.
simon	0:1014af42efd9	85	* There are separate instance structure declarations for each of the 3 supported data types.
simon	0:1014af42efd9	86	*
simon	0:1014af42efd9	87	* \par Initialization Functions
simon	0:1014af42efd9	88	* There is also an associated initialization function for each data type.
simon	0:1014af42efd9	89	* The initialization function performs the following operations:
simon	0:1014af42efd9	90	* - Sets the values of the internal structure fields.
simon	0:1014af42efd9	91	* - Zeros out the values in the state buffer.
simon	0:1014af42efd9	92	* - Checks to make sure that the size of the input is a multiple of the decimation factor.
simon	0:1014af42efd9	93	*
simon	0:1014af42efd9	94	* \par
simon	0:1014af42efd9	95	* Use of the initialization function is optional.
simon	0:1014af42efd9	96	* However, if the initialization function is used, then the instance structure cannot be placed into a const data section.
simon	0:1014af42efd9	97	* To place an instance structure into a const data section, the instance structure must be manually initialized.
simon	0:1014af42efd9	98	* The code below statically initializes each of the 3 different data type filter instance structures
simon	0:1014af42efd9	99	* <pre>
simon	0:1014af42efd9	100	*arm_fir_decimate_instance_f32 S = {M, numTaps, pCoeffs, pState};
simon	0:1014af42efd9	101	*arm_fir_decimate_instance_q31 S = {M, numTaps, pCoeffs, pState};
simon	0:1014af42efd9	102	*arm_fir_decimate_instance_q15 S = {M, numTaps, pCoeffs, pState};
simon	0:1014af42efd9	103	* </pre>
simon	0:1014af42efd9	104	* where <code>M</code> is the decimation factor; <code>numTaps</code> is the number of filter coefficients in the filter;
simon	0:1014af42efd9	105	* <code>pCoeffs</code> is the address of the coefficient buffer;
simon	0:1014af42efd9	106	* <code>pState</code> is the address of the state buffer.
simon	0:1014af42efd9	107	* Be sure to set the values in the state buffer to zeros when doing static initialization.
simon	0:1014af42efd9	108	*
simon	0:1014af42efd9	109	* \par Fixed-Point Behavior
simon	0:1014af42efd9	110	* Care must be taken when using the fixed-point versions of the FIR decimate filter functions.
simon	0:1014af42efd9	111	* In particular, the overflow and saturation behavior of the accumulator used in each function must be considered.
simon	0:1014af42efd9	112	* Refer to the function specific documentation below for usage guidelines.
simon	0:1014af42efd9	113	*/
simon	0:1014af42efd9	114
simon	0:1014af42efd9	115	/**
simon	0:1014af42efd9	116	* @addtogroup FIR_decimate
simon	0:1014af42efd9	117	* @{
simon	0:1014af42efd9	118	*/
simon	0:1014af42efd9	119
simon	0:1014af42efd9	120	/**
simon	0:1014af42efd9	121	* @brief Processing function for the floating-point FIR decimator.
simon	0:1014af42efd9	122	* @param[in] *S points to an instance of the floating-point FIR decimator structure.
simon	0:1014af42efd9	123	* @param[in] *pSrc points to the block of input data.
simon	0:1014af42efd9	124	* @param[out] *pDst points to the block of output data.
simon	0:1014af42efd9	125	* @param[in] blockSize number of input samples to process per call.
simon	0:1014af42efd9	126	* @return none.
simon	0:1014af42efd9	127	*/
simon	0:1014af42efd9	128
simon	0:1014af42efd9	129	void arm_fir_decimate_f32(
simon	0:1014af42efd9	130	const arm_fir_decimate_instance_f32 * S,
simon	0:1014af42efd9	131	float32_t * pSrc,
simon	0:1014af42efd9	132	float32_t * pDst,
simon	0:1014af42efd9	133	uint32_t blockSize)
simon	0:1014af42efd9	134	{
simon	0:1014af42efd9	135	float32_t pState = S->pState; / State pointer */
simon	0:1014af42efd9	136	float32_t pCoeffs = S->pCoeffs; / Coefficient pointer */
simon	0:1014af42efd9	137	float32_t pStateCurnt; / Points to the current sample of the state */
simon	0:1014af42efd9	138	float32_t px, pb; /* Temporary pointers for state and coefficient buffers */
simon	0:1014af42efd9	139	float32_t sum0; /* Accumulator */
simon	0:1014af42efd9	140	float32_t x0, c0; /* Temporary variables to hold state and coefficient values */
simon	0:1014af42efd9	141	uint32_t numTaps = S->numTaps; /* Number of filter coefficients in the filter */
simon	0:1014af42efd9	142	uint32_t i, tapCnt, blkCnt, outBlockSize = blockSize / S->M; /* Loop counters */
simon	0:1014af42efd9	143
simon	0:1014af42efd9	144	/* S->pState buffer contains previous frame (numTaps - 1) samples */
simon	0:1014af42efd9	145	/* pStateCurnt points to the location where the new input data should be written */
simon	0:1014af42efd9	146	pStateCurnt = S->pState + (numTaps - 1u);
simon	0:1014af42efd9	147
simon	0:1014af42efd9	148	/* Total number of output samples to be computed */
simon	0:1014af42efd9	149	blkCnt = outBlockSize;
simon	0:1014af42efd9	150
simon	0:1014af42efd9	151	while(blkCnt > 0u)
simon	0:1014af42efd9	152	{
simon	0:1014af42efd9	153	/* Copy decimation factor number of new input samples into the state buffer */
simon	0:1014af42efd9	154	i = S->M;
simon	0:1014af42efd9	155
simon	0:1014af42efd9	156	do
simon	0:1014af42efd9	157	{
simon	0:1014af42efd9	158	pStateCurnt++ = pSrc++;
simon	0:1014af42efd9	159
simon	0:1014af42efd9	160	} while(--i);
simon	0:1014af42efd9	161
simon	0:1014af42efd9	162	/* Set accumulator to zero */
simon	0:1014af42efd9	163	sum0 = 0.0f;
simon	0:1014af42efd9	164
simon	0:1014af42efd9	165	/* Initialize state pointer */
simon	0:1014af42efd9	166	px = pState;
simon	0:1014af42efd9	167
simon	0:1014af42efd9	168	/* Initialize coeff pointer */
simon	0:1014af42efd9	169	pb = pCoeffs;
simon	0:1014af42efd9	170
simon	0:1014af42efd9	171	/* Loop unrolling. Process 4 taps at a time. */
simon	0:1014af42efd9	172	tapCnt = numTaps >> 2;
simon	0:1014af42efd9	173
simon	0:1014af42efd9	174	/* Loop over the number of taps. Unroll by a factor of 4.
simon	0:1014af42efd9	175	** Repeat until we've computed numTaps-4 coefficients. */
simon	0:1014af42efd9	176	while(tapCnt > 0u)
simon	0:1014af42efd9	177	{
simon	0:1014af42efd9	178	/* Read the b[numTaps-1] coefficient */
simon	0:1014af42efd9	179	c0 = *(pb++);
simon	0:1014af42efd9	180
simon	0:1014af42efd9	181	/* Read x[n-numTaps-1] sample */
simon	0:1014af42efd9	182	x0 = *(px++);
simon	0:1014af42efd9	183
simon	0:1014af42efd9	184	/* Perform the multiply-accumulate */
simon	0:1014af42efd9	185	sum0 += x0 * c0;
simon	0:1014af42efd9	186
simon	0:1014af42efd9	187	/* Read the b[numTaps-2] coefficient */
simon	0:1014af42efd9	188	c0 = *(pb++);
simon	0:1014af42efd9	189
simon	0:1014af42efd9	190	/* Read x[n-numTaps-2] sample */
simon	0:1014af42efd9	191	x0 = *(px++);
simon	0:1014af42efd9	192
simon	0:1014af42efd9	193	/* Perform the multiply-accumulate */
simon	0:1014af42efd9	194	sum0 += x0 * c0;
simon	0:1014af42efd9	195
simon	0:1014af42efd9	196	/* Read the b[numTaps-3] coefficient */
simon	0:1014af42efd9	197	c0 = *(pb++);
simon	0:1014af42efd9	198
simon	0:1014af42efd9	199	/* Read x[n-numTaps-3] sample */
simon	0:1014af42efd9	200	x0 = *(px++);
simon	0:1014af42efd9	201
simon	0:1014af42efd9	202	/* Perform the multiply-accumulate */
simon	0:1014af42efd9	203	sum0 += x0 * c0;
simon	0:1014af42efd9	204
simon	0:1014af42efd9	205	/* Read the b[numTaps-4] coefficient */
simon	0:1014af42efd9	206	c0 = *(pb++);
simon	0:1014af42efd9	207
simon	0:1014af42efd9	208	/* Read x[n-numTaps-4] sample */
simon	0:1014af42efd9	209	x0 = *(px++);
simon	0:1014af42efd9	210
simon	0:1014af42efd9	211	/* Perform the multiply-accumulate */
simon	0:1014af42efd9	212	sum0 += x0 * c0;
simon	0:1014af42efd9	213
simon	0:1014af42efd9	214	/* Decrement the loop counter */
simon	0:1014af42efd9	215	tapCnt--;
simon	0:1014af42efd9	216	}
simon	0:1014af42efd9	217
simon	0:1014af42efd9	218	/* If the filter length is not a multiple of 4, compute the remaining filter taps */
simon	0:1014af42efd9	219	tapCnt = numTaps % 0x4u;
simon	0:1014af42efd9	220
simon	0:1014af42efd9	221	while(tapCnt > 0u)
simon	0:1014af42efd9	222	{
simon	0:1014af42efd9	223	/* Read coefficients */
simon	0:1014af42efd9	224	c0 = *(pb++);
simon	0:1014af42efd9	225
simon	0:1014af42efd9	226	/* Fetch 1 state variable */
simon	0:1014af42efd9	227	x0 = *(px++);
simon	0:1014af42efd9	228
simon	0:1014af42efd9	229	/* Perform the multiply-accumulate */
simon	0:1014af42efd9	230	sum0 += x0 * c0;
simon	0:1014af42efd9	231
simon	0:1014af42efd9	232	/* Decrement the loop counter */
simon	0:1014af42efd9	233	tapCnt--;
simon	0:1014af42efd9	234	}
simon	0:1014af42efd9	235
simon	0:1014af42efd9	236	/* Advance the state pointer by the decimation factor
simon	0:1014af42efd9	237	* to process the next group of decimation factor number samples */
simon	0:1014af42efd9	238	pState = pState + S->M;
simon	0:1014af42efd9	239
simon	0:1014af42efd9	240	/* The result is in the accumulator, store in the destination buffer. */
simon	0:1014af42efd9	241	*pDst++ = sum0;
simon	0:1014af42efd9	242
simon	0:1014af42efd9	243	/* Decrement the loop counter */
simon	0:1014af42efd9	244	blkCnt--;
simon	0:1014af42efd9	245	}
simon	0:1014af42efd9	246
simon	0:1014af42efd9	247	/* Processing is complete.
simon	0:1014af42efd9	248	** Now copy the last numTaps - 1 samples to the satrt of the state buffer.
simon	0:1014af42efd9	249	** This prepares the state buffer for the next function call. */
simon	0:1014af42efd9	250
simon	0:1014af42efd9	251	/* Points to the start of the state buffer */
simon	0:1014af42efd9	252	pStateCurnt = S->pState;
simon	0:1014af42efd9	253
simon	0:1014af42efd9	254	i = (numTaps - 1u) >> 2;
simon	0:1014af42efd9	255
simon	0:1014af42efd9	256	/* copy data */
simon	0:1014af42efd9	257	while(i > 0u)
simon	0:1014af42efd9	258	{
simon	0:1014af42efd9	259	pStateCurnt++ = pState++;
simon	0:1014af42efd9	260	pStateCurnt++ = pState++;
simon	0:1014af42efd9	261	pStateCurnt++ = pState++;
simon	0:1014af42efd9	262	pStateCurnt++ = pState++;
simon	0:1014af42efd9	263
simon	0:1014af42efd9	264	/* Decrement the loop counter */
simon	0:1014af42efd9	265	i--;
simon	0:1014af42efd9	266	}
simon	0:1014af42efd9	267
simon	0:1014af42efd9	268	i = (numTaps - 1u) % 0x04u;
simon	0:1014af42efd9	269
simon	0:1014af42efd9	270	/* copy data */
simon	0:1014af42efd9	271	while(i > 0u)
simon	0:1014af42efd9	272	{
simon	0:1014af42efd9	273	pStateCurnt++ = pState++;
simon	0:1014af42efd9	274
simon	0:1014af42efd9	275	/* Decrement the loop counter */
simon	0:1014af42efd9	276	i--;
simon	0:1014af42efd9	277	}
simon	0:1014af42efd9	278	}
simon	0:1014af42efd9	279
simon	0:1014af42efd9	280	/**
simon	0:1014af42efd9	281	* @} end of FIR_decimate group
simon	0:1014af42efd9	282	*/

Repository toolbox

Export to desktop IDE

Repository details

Type:	Library
Created:	10 Mar 2011
Imports:	907
Forks:	1
Commits:	3
Dependents:	5
Dependencies:	0
Followers:	35

src/Cortex-M4-M3/FilteringFunctions/arm_fir_decimate_f32.c@0:1014af42efd9, 2011-03-10 (annotated)

Who changed what in which revision?

Repository toolbox

Repository details

Important Information for this Arm website

Access Warning