Thanks Richard,
just wanted to understand it. Now it is clear.
|
Re: Heart rate sensor MAX30102
Thanks Richard,
just wanted to understand it. Now it is clear.
|
Re: Heart rate sensor MAX30102
now have low pass filtering on pic
73 rpm tonight
Warning I'm not a teacher
|
Re: Heart rate sensor MAX30102
That's really impressed Richard.
Im still working step by step on the code, im still at an early stage.
Just need to understand the methodology.
|
Re: Heart rate sensor MAX30102
try as i might i cannot detect the relevant peaks reliably
Code:'****************************************************************'* Name : MAX30102 FOR PIC 18 DEMO * '* Author : Richard * '* Notice : * '* : * '* Date : 9/1/2022 * '* Version : 1.0 * '* Notes : MAX30102 * '* :18f26k22 @64Mhz * '**************************************************************** #CONFIG ; The PBP configuration for the PIC18F26K22 is: CONFIG FOSC = INTIO67 CONFIG PLLCFG = OFF ;Oscillator multiplied by 4 CONFIG PRICLKEN = ON ;Primary clock enabled CONFIG FCMEN = OFF ;Fail-Safe Clock Monitor disabled CONFIG IESO = OFF ;Oscillator Switchover mode disabled CONFIG BOREN = SBORDIS ; Brown-out Reset enabled in hardware only (SBOREN is disabled) CONFIG WDTEN = ON ; WDT is always enabled. SWDTEN bit has no effect ;| CONFIG WDTPS = 32768 ; 1:32768 ---> HERE enable the watchdog timer with a 1:32768 postscale;| CONFIG PWRTEN = ON CONFIG HFOFST = ON ; HFINTOSC output and ready status are not delayed by the oscillator stable status CONFIG MCLRE = EXTMCLR ; MCLR pin enabled, RE3 input pin disabled CONFIG LVP = OFF ; Single-Supply ICSP disabled CONFIG XINST = OFF ; Instruction set extension and Indexed Addressing mode disabled (Legacy mode) CONFIG DEBUG = OFF ; Disabled CONFIG CP0 = OFF ; Block 0 (000800-003FFFh) not code-protected CONFIG CP1 = OFF ; Block 1 (004000-007FFFh) not code-protected CONFIG CP2 = OFF ; Block 2 (008000-00BFFFh) not code-protected CONFIG CP3 = OFF ; Block 3 (00C000-00FFFFh) not code-protected CONFIG CPB = OFF ; Boot block (000000-0007FFh) not code-protected CONFIG CPD = OFF ; Data EEPROM not code-protected CONFIG WRT0 = OFF ; Block 0 (000800-003FFFh) not write-protected CONFIG WRT1 = OFF ; Block 1 (004000-007FFFh) not write-protected CONFIG WRT2 = OFF ; Block 2 (008000-00BFFFh) not write-protected CONFIG WRT3 = OFF ; Block 3 (00C000-00FFFFh) not write-protected CONFIG WRTC = OFF ; Configuration registers (300000-3000FFh) not write-protected CONFIG WRTB = OFF ; Boot Block (000000-0007FFh) not write-protected CONFIG WRTD = OFF ; Data EEPROM not write-protected CONFIG EBTR0 = OFF ; Block 0 (000800-003FFFh) not protected from table reads executed in other blocks CONFIG EBTR1 = OFF ; Block 1 (004000-007FFFh) not protected from table reads executed in other blocks CONFIG EBTR2 = OFF ; Block 2 (008000-00BFFFh) not protected from table reads executed in other blocks CONFIG EBTR3 = OFF ; Block 3 (00C000-00FFFFh) not protected from table reads executed in other blocks CONFIG EBTRB = OFF ; Boot Block (000000-0007FFh) not protected from table reads executed in other blocks #ENDCONFIG goto overasm asm PutMulResult?D macro Din MOVE?BB Din, R2 MOVE?BB Din + 1, R2 + 1 MOVE?BB Din + 2, R0 MOVE?BB Din + 3, R0 + 1 RST?RP endm endasm overasm: define OSC 64 OSCCON = %01110000 ; 16Mhz OSCTUNE.6 = 1 ; Enable 4x PLL while ! osccon2.7 :WEND ; to make sure the pll has stabilised before you run any other code TRISC = %11011000 ANSELC=0 ANSELB=0 #DEFINE DEBUGING 1 DEFINE DEBUG_REG PORTB DEFINE DEBUG_BIT 7 DEFINE DEBUG_BAUD 9600 DEFINE DEBUG_MODE 0 SDA var portc.4 SCL VAR portc.3 reg var BYTE fifo var byte[6] bank0 addr var byte REVISION var byte PARTID var byte buffer var word[384] ptr var byte btr var byte index var word average var byte[4] bank0 ave var word peaks var byte[5] lastpeak var word lpfdata var word[192] result var word ext result1 var word ext @result = _fifo @result1 = _fifo + 3 ;----------------------- REGISTER ------------------------- REG_TEMP_INT con $1F REG_TEMP_FRAC con $20 REG_TEMP_CONFIG con $21 REG_PART_REVISION con $FE REG_MODE con $09 REG_SPO2_CONFIG con $0A ;SPO2_ADC_RGE[6:5]SPO2_SR[4:2]LED_PW[1:0] REG_LED1_PA con $0C REG_LED2_PA con $0D FIFO_WR_PTR con $04 ;FIFO_WR_PTR[4:0] FIFO_RD_PTR con $06 FIFO_DATA con $07 FIFO_CONFIG con $08 ;FIFO_A_FULL[3:0] SMP_AVE[7:5] OVF_COUNTER con $05 STATUS1 con 0 ;A_FULL[7] INTERRUPT_EN con 2 ' -----------------------------------------------------------------------------| #IFDEF DEBUGING LATB.7=1 PAUSE 2000 DEBUG 13,10, "READY" #ENDIF ADDR = $ae ; addr is 0x57 << 1 reg = REG_PART_REVISION i2cread SDA,scl,ADDR,reg,[REVISION,PARTID] reg = REG_MODE i2cwrite SDA,scl,ADDR,reg,[3] reg = REG_SPO2_CONFIG i2cwrite SDA,scl,ADDR,reg,[$20] reg = REG_LED2_PA i2cwrite SDA,scl,ADDR,reg,[10] reg = REG_LED1_PA i2cwrite SDA,scl,ADDR,reg,[10] reg = FIFO_CONFIG i2cwrite SDA,scl,ADDR,reg,[16] reg = INTERRUPT_EN i2cwrite SDA,scl,ADDR,reg,[$E0] DEBUG 13,10, "PART ID: 0X",hex PARTID,"/0X",hex REVISION start: reg= REG_TEMP_CONFIG i2cwrite SDA,scl,ADDR,reg,[1] reg= REG_TEMP_INT i2cread sda,scl,ADDR,reg,[fifo[0],fifo[1]] DEBUG 13,10, "TEMP: ",DEC fifo[0],".",DEC2( fifo[1]<<6) GOSUB AQUIRE pause 2000 goto start aquire: reg= FIFO_WR_PTR i2cwrite SDA,scl,ADDR,reg,[0] reg= FIFO_RD_PTR i2cwrite SDA,scl,ADDR,reg,[0] reg = OVF_COUNTER i2cwrite sda,scl,ADDR,reg,[0] asm banksel 0 clrf _average clrf _average+1 clrf _average+2 clrf _average+3 endasm FOR index = 0 TO 255 STEP 0 btr = 0 reg = STATUS1 WHILE (btr & 192) = 0 i2cread sda,scl,ADDR,reg,[btr] WEND reg = FIFO_WR_PTR i2cread SDA,scl,ADDR,reg,[Ptr] reg = FIFO_RD_PTR i2cread SDA,scl,ADDR,reg,[btr] btr = (ptr - btr) & 31 WHILE btr reg = FIFO_DATA ;---------ir----------;===========red=========== i2cread sda,scl,ADDR,reg,[FIFO[5],FIFO[4],FIFO[3],FIFO[2],FIFO[1],FIFO[0]] asm ;ROTATE READING TO GET SUFFICIENT RESOLOUTION INTO lower WORD banksel 0 bcf STATUS,C RRCF _fifo + 2 ,f RRCF _fifo + 1 ,f RRCF _fifo ,f bcf STATUS,C RRCF _fifo + 2 ,f RRCF _fifo + 1 ,f RRCF _fifo ,f bcf STATUS,C RRCF _fifo + 5 ,f RRCF _fifo + 4 ,f RRCF _fifo + 3 ,f bcf STATUS,C RRCF _fifo + 5 ,f RRCF _fifo + 4 ,f RRCF _fifo + 3 ,f movf _fifo ,w ;RED AVERAGE addwf _average,f movf _fifo + 1 ,w bnc s1 incf _average + 1,f s1 addwf _average + 1,f movf _fifo + 2 ,w bnc s2 incf _average + 2,f s2 bnc s3 incf _average + 3 ,f s3 endasm buffer[index] = result ;red index = index + 1 buffer[index] = result1 ;ir index = index + 1 btr = btr - 1 WEND NEXT DEBUG 13,10,hex2 average[3],hex2 average[2],hex2 average[1],hex2 average @ PutMulResult?D _average ;RED AVERAGE ave = DIV32 128 ;RED AVERAGE reg = OVF_COUNTER i2cread sda,scl,ADDR,reg,[reg ] DEBUG 13,10,"OverFlows ",DEC reg,9,dec ave,9,hex ave FOR index = 0 TO 255 STEP 0 ave = (ave ** 62259) + (buffer[index] ** 3276 ) lpfdata[index>>1]=ave ;.....lpf..;;;;----red-----;;;;======-ir-======== DEBUG 13,10,dec ave,9,dec buffer[index],9,dec buffer[index+1] index = index + 2 NEXT RETURN
Warning I'm not a teacher
|
Re: Heart rate sensor MAX30102
In Post #44 your low-pass filter seemed to work perfectly, then in Post #49, it seemed a bit overdone. Then in Post #51 it looks like the low-pass filter is acting like a 500F capacitor! I didn't look at the code, but maybe you could back off on the filter settings for a clearer reading.
|
Re: Heart rate sensor MAX30102
i have tried many variations , the added aggressiveness of the filter is an attempt to remove the unwanted "little" peaksbut maybe you could back off on the filter settings for a clearer reading.
what looks like a nice picture of on a graph does not mean peak detection is easier.
some of the lp filtering is just added in with excel also, even tried using trend lines in attempt at detecting pos and neg zero crossings
that's still not 100% effective. dsp is not my forte, dc drift in the readings can be neg or pos going or both or zero.
fft was a failure because i cannot remove the dc component effectively. i need a better method i have no effective strategy
Warning I'm not a teacher
|
Re: Heart rate sensor MAX30102
Here is a Filter algorithm that has served me well with several projects. For the moment I'm going to assume your code serves 2 main functions: read the heart rate MAX30102 and report the results. This means the processor execution time can be maximized for these 2 functions.
Reporting process is what it is. That's whatever protocol (UART, SPI, I2C, or LCDOUT etc) you have chosen to make use of the processed data.
As for reading and filtering the MAX30102 data, the first stage is to take multiple ADC reads (or appropriately retrieve the latest data report) each Subroutine CALL and average them. The result gets fed into a circular buffer, where the average is the last X readings, and what you shoot out in your report. It looks something like this:
Somewhere a CALL (GOSUB) is made to check the input from the MAX30102 sensor. Within the Subroutine, you do this 4X and average the results. The average result is fed into a [4] BYTE/WORD circular buffer. An Accumulator adds all buffer entries and divides the result by the number of entries (calculates an average). The value reported is the average of the circular buffer -- buffered value. With me so far? Let's take it for a test drive, shall we?
To change the impact of the filter, you can increase or decrease the size of the HrValBuf[4] buffer and/or the HrValCir[4] buffer. If too jumpy, add more entries for either/both buffer(s). Conversely, to economize, try reducing buffer sizes and see if you still like the results. Changes would have to be reflected in Get_Heartrate:. No, of course this isn't the absolute most efficient code to accomplish the goal, and it is too abbreviated to actually work, but I hope it illustrates an effective filter that has served me well over many projects.Code:b0 VAR BYTE ;Used for FOR/LOOP HrVal VAR BYTE ;if using 8-bit ADC, ...VAR WORD for 10-bit, or whatever protocol demands HrValBuf VAR BYTE[4] ;again, WORD for 10-bit, this is your 4 immediate ADC (MAX30102 input) reads Buffer HrValCir VAR BYTE[4] ;again, WORD for 10-bit, this is your 4 entry Circular Buffer HrBufAcc VAR WORD ;Total of all 4 HrValBuf readings HrCirAcc Var WORD ;Total of all 4 HrCirBuf readings HeartRate VAR BYTE ;again, WORD for 10-bit, End Filtered Result to be Reported or Processed ... ;from Somewhere in Code: GOSUB Get_HeartRate GOSUB Send_HeartRate ;Do something with the HeartRate Value Acquired & Filtered ... ;then Continue Code... Get_Heartrate: HrValCir[3] = HrValCir[2] ;Rotate the Circular Buffer HrValCir[2] = HrValCir[1] HrValCir[1] = HrValCir[0] FOR b0 = 0 TO 3 ADCIN MAX30102,HrVal ;or I2CREAD or SPIREAD -- Retrieve a Current Value HrValBuf[b0] = HrVal NEXT b0 HrBufAcc = HrValBuf[3] + HrValBuf[2] + HrValBuf[1] + HrValBuf[0] ;Add All 4 Readings... HrValCir[0] = HrBufAcc >> 2 ;Same as HrValCir[0] = HrBufAcc / 4 -- Places the Latest Averaged Value into the Circular Buffer HrCirAcc = HrValCir[3] + HrValCir[2] + HrValCir[1] + HrValCir[0] ;Totals the Circular Buffer HeartRate = HrCirAcc >> 2 ;Same as HeartRate = HrCirAcc / 4, Averages the Circular Buffer RETURN
Last edited by mpgmike; - 15th January 2022 at 01:52.
|
Re: Heart rate sensor MAX30102
been there done that.I think Richard is eluding to a process called Fast Fourier Transform
Originally Posted by richard
there is plenty of raw data posted, you can always ask for more. feel free to have a goThe one that is like Example #1 in the textbooks is simply:
Last edited by richard; - 19th January 2022 at 22:22.
Warning I'm not a teacher
Members who have read this thread : 0You do not have permission to view the list of names.
Posting Permissions
Bookmarks