Timing Constants
By default, we use a 15ms timeslot template:
| common | GINA | TelosB | |||||
|
| provisioned | measured | provisioned | measured | ||
Time-slot related |
|
|
|
|
| ||
4000us |
|
|
|
| |||
1300us |
|
|
|
| |||
4000us |
|
|
|
| |||
500us |
|
|
|
| |||
15000us |
|
|
|
| |||
Execution speed related |
|
|
|
|
| ||
| 2014us | 746us | 2899us | 2420us | |||
| 305us | 83us | 610us | 474us | |||
| 1007us | 84us | 1000us | 477us | |||
| 305us | 219us | 793us | 746us | |||
Radio speed related |
|
|
|
|
| ||
| 214us | 219us | 366us | 352us | |||
| 0us | cannot | 0us | cannot | |||
Radio watchdog |
|
|
|
|
| ||
1000us |
|
|
|
| |||
5000us |
|
|
|
| |||
3000us |
|
|
|
| |||
...
The image below shows TsTxOffset, measured between the beginning of the slot and the time the receiver gets the start of frame event from the radio.
(source file at nine_slots_final.prj)
...
How early the receiver starts listening for a data packet, see TschSynchronization.
(source file at nine_slots_final.prj)
...
The duration between the end of the data packet, and the beginning of the ACK packet (its SFD).
(source file at nine_slots_final.prj)
...
Exactly the same use as #TsLongGT, but when listening for an ACK. Because this duration is measured from the end of the data packet, which just happened, motes are "very" synchronized and so TsShortGT can be shorted than #TsLongGT.
(source file at nine_slots_final.prj)
...
Determined by the pre-agreed upon timeslot template.
The duration of a slot.
(source file at nine_slots_final.prj)
...
This is the maximum time it takes your microcontroller to execute #ti2 and #ta1. You want maxTxDataPrepare to be as close as possible to the measured value to reduce the energy wasted while in the TXDATAREADY state.
(source file at nine_slots_final.prj)
...
This is the maximum time it takes your microcontroller to execute #ti6. You want maxRxAckPrepare to be as close as possible to the measured value to reduce the energy wasted while in the RXACKREADY state.
(source file at nine_slots_final.prj)
...
This is the maximum time it takes your microcontroller to execute #ri2. You want maxRxDataPrepare to be as close as possible to the measured value to reduce the energy wasted while in the TRXDATAREADY state.
(source file at nine_slots_final.prj)
...
This is the maximum time it takes your microcontroller to execute #ri6 and #ra2. You want maxTxAckPrepare to be as close as possible to the measured value to reduce the energy wasted while in the TXACKREADY state.
(source file at nine_slots_final.prj)
...
The best way to determine this value is by measuring. If your radio does not generate an event when it has sent the SFD, you can measure the SFD receive event:
(source file at nine_slots_final.prj)
...
This should be longer than the duration of the longest packet. We measure the duration of the longest packet (127 byte payload) to be 4.27ms. The value of wdDataDuration simply needs to be longer than that.
(source file at nine_slots_final.prj)
...
Similar to #wdDataDuration, but for the ACK. The ACK is much shorter and we measure the time it takes to transmit it to be 578us.
The value for wdAckDuration simply needs to be longer than that.
(source file at nine_slots_final.prj)