 |
 |
| View previous topic :: View next topic |
| Author |
Message |
Andrew83
Joined: 16 Sep 2008
Posts: 51
|
| IR decoding question |
 Posted: Sun Oct 26, 2008 5:17 am |
 |
|
Hello 2 all !
I'm trying to decode the "header" part of a REC80 IR protocol.
This header contains a 9 ms low portion and a 4,5 ms high portion.
An example is illustrated in the code below.
My problem is that although I am able to succesfully recognize the first 9 ms low portion of the transmission, I can't recognize the second part (the 4,5 ms high).
I am using a pic18f452, with a 4.4 MHZ oscillator and using CCS C version 4.057.
I think that somewhere is a trivial mistake but I can't figure it out !! 
P.S. The example code was provided by this post :http://www.ccsinfo.com/forum/viewtopic.php?t=36124&highlight=wireless
THANK YOU !!
| Code: |
//This is a test program for REC80 IR protocol.
//Trying to determine the "header" part of a IR transmission.
//Transmision example:
//------\____________/------------\__/--\__/--\__/--\__/--\__/--\__
//(idle)| header low header high | the rest of the transmission |
// (9 ms) (4,5 ms)
//Hardware -
// TSOP1738 on Pin B0
// TSOP data is inverted; it idles high.
// [--\_ : negative going edge; _/-- : positive going edge]
//
// Original author :Rohit de Sa
// Modifyed by : Andrew83
#include <18F452.h>
#fuses HS,NOPROTECT,NOOSCSEN,BROWNOUT,BORV45,NOWDT,WDT128,PUT,NOSTVREN,NODEBUG,NOLVP,WRT,NOCPB,WRTB,WRTC,NOCPD,NOWRTD,NOEBTR,NOEBTRB
#use delay(clock=4.4MHZ)
#use rs232(baud=9600, xmit=PIN_C6, rcv=PIN_C7)
#use fast_io(b)
#define start0_min 8000 //no of counts to safely detect a "0" start
#define start0_max 12000 //optimal @1,1 MIPS is 9900
#define start1_min 3000 //no of counts to safely detect a "1" start
#define start1_max 6000 //optimal @1,1 MIPS is 4950
#define tsop pin_b0
int16 start_address=0; //ROM address of 'Start' label
#int_timer1 //if timeout error
void timer1_isr()
{
*0xffd=start_address; //modify return address in TOS
*0xffe=start_address/256;
}
void main()
{
int16 timer_value=0;
//int1 irdone=false;
delay_ms(100); //setting up PIC
setup_adc_ports(no_analogs);
setup_adc(adc_off);
set_tris_b(0b00000001);
start_address=label_address(start_ir); //ROM address of IR detection routine
delay_ms(100);
setup_timer_1(t1_internal|t1_div_by_1);
disable_interrupts(int_timer1);
enable_interrupts(global);
while(1)
{
start_ir:
disable_interrupts(int_timer1);
clear_interrupt(int_timer1);
while(1)
{
while(input(tsop)); //polling the input RB0
set_timer1(0); //set timer1 to zero
enable_interrupts(int_timer1); //enable timer1 interrupt
delay_us(10); //settling delay
while(!(input(tsop))); //fall of signal detected
timer_value = get_timer1(); //get timer1 value
set_timer1(0); //reset timer1 to zero
if (!((start0_min < timer_value) && (timer_value < start0_max))) //compare values to see if it's a valid 9 ms "zero" duration
{
printf("\nNo first part of signal header !"); //first part of header not detected
//break;
}
else
{ printf("\nDetected a 0 pulse value:%lu",timer_value); } // print value for "0" duration
delay_us(10); //settling delay
while(input(tsop)); // if the input RB0 is high again
timer_value = get_timer1(); //get timer1 value
set_timer1(0); //reset timer1 to zero
if (!((start1_min < timer_value) && (timer_value < start1_max))) //compare values to see if it's a valid 4,5 ms "one" duration
{
printf("\nNo second part of signal header !"); //second part of header not detected
//break;
}
else
{ printf("\nDetected a 1 pulse value:%lu",timer_value); } // print value for "1" duration
}
}
}
|
|
|
 |
Andrew83
Joined: 16 Sep 2008
Posts: 51
|
|
| Posted: Sun Oct 26, 2008 5:38 am |
|
|
Just a little add-on to what i've stated.
The results i'm getting from my test are :
- for a low duration (9 ms) a counter value of 10057 -> almost spot on
- for a high duration (4,5 ms) a counter value of about 33681 ->this is where the problem lies : it shold be much much less.
According to my calculations, the counter value should be around 4950 or so.  |
|
|
ckielstra
Joined: 18 Mar 2004
Posts: 3680
Location: The Netherlands
|
|
| Posted: Sun Oct 26, 2008 7:18 am |
|
|
| Code: | | { printf("\nDetected a 0 pulse value:%lu",timer_value); } |
At 9600 baud it takes about 1ms to transmit each character.
+/- 30 characters = 30ms = 33000 timer1 ticks, very close to your measured value.
Instead of sending a text try setting one or more LEDs for debugging info. |
|
|
Andrew83
Joined: 16 Sep 2008
Posts: 51
|
|
| Posted: Mon Oct 27, 2008 1:20 am |
|
|
Thank you ckielstra  ! ..I will try this solution and let you know ! |
|
|
FvM
Joined: 27 Aug 2008
Posts: 2337
Location: Germany
|
|
| Posted: Mon Oct 27, 2008 1:37 am |
|
|
As pointed out, directly sending text through RS232 during time critical program section can't work.
If the required information is mainly in the measured time value, you can also store it in a variable and
print it after finishing the timing measurement.
Another general solution is by operating RS232 output buffered and interrupt driven through a FIFO. |
|
|
Andrew83
Joined: 16 Sep 2008
Posts: 51
|
|
| Posted: Mon Oct 27, 2008 3:14 am |
|
|
Works flawless ! Thak you again for your quick answers ! This field is very new to me, but I am willing to learn.
| Quote: | | Another general solution is by operating RS232 output buffered and interrupt driven through a FIFO. |
I would like to know more about how this method could be implemented. |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
