CCS/MSP432P401R: MSP432 uses CMSIS function library and arm_fir_q15 function. In this function filter, why is the result of input signal? Has anyone ever used this function? Can you enlighten me? Is there a routine of this function? Thanks

Hello,

I apologize but I did not understand the question. Are you asking us where the output of "arm_fir_q15" function is stored? I haven't used CMSIS DSPLib so far, but does the documentation of the CMSIS DSPLib not mention this information?

Thanks,
Sai

Do you have a routine for this function? Can I have one, please? Thanks

Do you have a routine for this function? Can I have one, please? Thanksc

Hello Forum User,

Did you check the link I sent earlier and the associated example provided by ARM?

I can call the arm_fir_f32 function, but I still can't call the arm_fir_q15 program correctly.

Hello Forum User,

Can you please post a code snippet and using the snippet elaborate on what is not working?

this is my code.
#include "driverlib.h"
#include <stdint.h>
#include <stdbool.h>
#include "arm_math.h"
#include "arm_const_structs.h"
#define TEST_LENGTH_SAMPLES 320
#define BLOCK_SIZE 320
#define NUM_TAPS 28
uint32_t blockSize = BLOCK_SIZE;
uint32_t numBlocks = TEST_LENGTH_SAMPLES/BLOCK_SIZE; /
static int16_t testInput_q15_50Hz_200Hz[TEST_LENGTH_SAMPLES]={0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40,0,40,38,7,0,32,61,45,0,-21,0,21,0,-45,-61,-32,0,-7,-38,-40}; /* 采样点 */
static int16_t testOutput[TEST_LENGTH_SAMPLES];
static int16_t firStateF32[BLOCK_SIZE + NUM_TAPS -1];
int16_t firCoeffs32LP[NUM_TAPS] = {0,0,0,1,1,1,-1,-3,-5,-3,4,14,24,31,31,24,14,4,-3,-5,-3,-1,1,1,1,0,0,0};
static void arm_fir_q15_lp(void)
{
arm_fir_instance_q15 S;
arm_fir_init_q15(&S,NUM_TAPS, &firCoeffs32LP, &firStateF32, blockSize);

arm_fir_q15(&S,&testInput_q15_50Hz_200Hz,&testOutput, blockSize);
}
void main(void)
{
/* Stop Watchdog */
MAP_WDT_A_holdTimer();
arm_fir_q15_lp();
while(1);
}
Low frequency filtering, fs=1000Hz, fc=125Hz for sinusoidal signal with frequency of 50Hz and 200Hz.
The output is wrong, and it doesn't play the role of filtering.

Thank you very much.

Hello Forum User

From where did you generate the Coefficients for the FIR? I use MatLab program to generate the Tap values and I do not see an issue. Also if you look at the example in the link I sent to you, the blocks have to be run through the FIR filter. In your case you have not done anything like that. Please check the URL I sent you to first correct the code.

Hello Forum user

Was the issue resolved?