Skip to content
This repository has been archived by the owner on Dec 14, 2021. It is now read-only.

Fix for join issues mentioned in "EU RX2 parameters #155", … #176

Merged
merged 2 commits into from
Jun 7, 2016

Conversation

kersing
Copy link
Contributor

@kersing kersing commented Jun 5, 2016

Modified the code to make sure JoinAccepts in RX2 get transmitted using SF12 as per LoRaWAN standard. Did not change scoreComputer to schedule response packets for SF7-SF11 in RX1 as it does not make sense given Thomas's remark concerning duty cycles of those channels.

As this patch does not fix the mentioned LMiC issues nor the requirement to set RX2 to SF9 for ABP nodes.

@jverhoeven
Copy link

Thanks for making the fix. Just tested it and my OTA join requests work on my mDot based node.

@htdvisser
Copy link
Contributor

Looks good, @kersing!

It would be even more awesome if you could also change this in the feature/us-frequencies branch (which runs on staging). Would you have time to do this as well?

@htdvisser htdvisser self-assigned this Jun 6, 2016
@htdvisser htdvisser merged commit 9b9f36b into TheThingsArchive:develop Jun 7, 2016
@tftelkamp
Copy link

I tested it, and it works fine.

However, the logic is broken now: we answer (SF9-SF11) join requests with a lower data rate, hence more airtime.

It would be much better if SF7-SF11 join accepts get responses in RX1, if duty-cycle allows.

For RX2 downlink message (at SF9), the logic is unchanged.

@jverhoeven
Copy link

Now that you mention it, I always get an answer in RX2. What is the reason I never get any responses in RX1? I run my own (multi-tech) gateway with hardly any users, so I would say there should be sufficient bandwidth on RX1.

@tftelkamp
Copy link

It mostly depends on the join request. An SF7 join will most likely get a response in RX1, unless the duty cycle limit has been reached.

On 09 Jun 2016, at 12:19, jverhoeven [email protected] wrote:

Now that you mention it, I always get an answer in RX2. What is the reason I never get any responses in RX1? I run my own (multi-tech) gateway with hardly any users, so I would say there should be sufficient bandwidth on RX1.


You are receiving this because you commented.
Reply to this email directly, view it on GitHub #176 (comment), or mute the thread https://github.com/notifications/unsubscribe/ANijYezSJjbBymxoc9ByiGYGT3nGC5zUks5qJ-i4gaJpZM4IudKZ.

@kersing
Copy link
Contributor Author

kersing commented Jun 9, 2016

@jverhoeven In my tests with MultiTech gateway (with polyfwd) join requests using SF7 and SF8 get response in RX1.

@kersing
Copy link
Contributor Author

kersing commented Jun 9, 2016

@tftelkamp Hmm, done some testing and using SF10 & SF11 with RX1 might not be a good idea. Got lots of denied messages because gateway could not transmit scheduled join-accept because of duty cycle limit.
Accept @ SF11 takes almost 750 ms which means a block for 75 secs. Using RX2 blocks that frequency for just 15 seconds.

BTW. shouldn't the dutycycle manager in the back end avoid using RX1 for the next 75 seconds??

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants