Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802.3_10SPE] Interpretation SEND_BEACON/SYNCING Clause 148



Hi Philip,
I believe that technically speaking you're correct. If your implementation allows for a large delay between TX and RX you may end up in such a strange situation where you enter the syncing state and skip it.
Indeed the intended behavior is that you resume cycling on transmit opportunities once you have detected the end of the BEACON.

To be honest I'm a bit unsure whether this is really something to be fixed in the state diagram or it is really an implementation issue.
If you wish to propose a change in D2.3 my personal advice would be to:
- Add a new state WAIT_BEACON, below SEND_BEACON.
- From SEND_BEACON enter this new state when beacon_timer_done, instead of SYNCING.
- Within WAIT_BEACON, set tx_cmd <= NONE.  // This ensures the duration of the BEACON is exactly beacon_timer.
- From WAIT_BEACON enter SYNCING state when rx_cmd = BEACON. // This should fix your issue
- Remove tx_cmd <= NONE from SYNCING state.
 
Kind Regards,
Piergiorgio

------ Original Message ------
From: "Philip Axer" <philip.axer@xxxxxxx>
To: STDS-802-3-10SPE@xxxxxxxxxxxxxxxxx
Sent: 6/2/2019 16:55:43
Subject: [802.3_10SPE] Interpretation SEND_BEACON/SYNCING Clause 148

Hi All, Hi Piergiorgio,

 

I have an interpretation question regarding the exit condition from SEND_BEACON. Specifically the interoperation between SEND_BEACON and SYNCING.

Let’s assume there is a large delay between feeding tx_cmd = BEACON until it is received back as rx_cmd. Let’s also assume local_nodeID=0 is in SEND_BEACON state and BEACON_TIMER has expired. So the BEACON was transmitted, but is not yet received back. That would mean that the SYNCING state is entered with rx_cmd!=BEACON (because it is not there yet), so SYNCING is more or less skipped. Is this an issue because the beacon will confuse the PHY and force it into EARLY_RECEIVE and eventually RECOVER?

 

I think what is intended is that you enter SYNCING, while still receiving the BEACON and hence consuming it in the SYNCING state.

 

Regards

Philip Axer


To unsubscribe from the STDS-802-3-10SPE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-10SPE&A=1


To unsubscribe from the STDS-802-3-10SPE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-10SPE&A=1