Timeline for Multiplexing serial line
Current License: CC BY-SA 3.0
5 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Mar 23, 2012 at 17:07 | history | edited | joeforker | CC BY-SA 3.0 |
added 6 characters in body
|
Mar 20, 2012 at 20:57 | history | edited | joeforker | CC BY-SA 3.0 |
added 16 characters in body
|
Mar 20, 2012 at 20:57 | comment | added | joeforker | The sensors each know their unique ID already. It doesn't have to matter if the micro misses a sync pulse. Configure the micro's 16-bit timer to have the same period as the sync pulse, reset the timer on a sync pulse, and wait for the timer to equal (id number) * (timer ticks per timeslot) before sending. | |
Mar 20, 2012 at 20:44 | comment | added | supercat | There is some advantage to having all sensors' processors running the same code; there is also some advantage to having everything sit on the same wires (rather than e.g. using a 74HC4017 to strobe each sensor in sequence). One complication I see for time-division multiplexing, though, is that I'm not sure how much the sensor CPU can do while it's waiting for a car. If one were to use TDM, it would more likely wait for a car, then wait for the next sync pulse, and then send the data. If desired, the data wire could easily handle the sync pulse as well if it were longer than the data. | |
Mar 20, 2012 at 18:18 | history | answered | joeforker | CC BY-SA 3.0 |