I’m using a RAK831 Pilot Gateway and ESP32 LoRaWAN (v2) nodes. I just ran my device overnight to make sure the problem was consistent so I’m convinced it must be a setting issue and not a signal one. At 60 second intervals, ABP mode, the LoRa Server on the RAK831 Pilot Gateway receives 8 packets in a row, then loses 8 packets in a row, and continues to follow that exact pattern indefinitely. Is there a particular file that contains the settings I might want to check? The RAK831 is running debian stretch with all lora server gateway/server components installed. The ESP32 node is using a modified LMIC library. I’m using US915.
What channels do you have enabled on your node?
Hopefully you have serial debug output from your node firmware. Make it print out the frequency it is transmitting on each time, make sure those are only frequencies to which your gateway is configured, and not say a set of 16 channels only 8 of which can be covered by a typical gateway.
The relevant code in the .ini file regarding the frequencies is:
#elif defined(CFG_us915)
// NA-US channels 0-71 are configured automatically.
// but only one group of 8 should (a subband) be active
// TTN recommends the second sub band, 1 in a zero based count.
// https://github.com/TheThingsNetwork/gateway-conf/blob/master/US-global_conf.json
LMIC_selectSubBand(1);
I do have a serial debug output, for TXMODE it goes like this:
TXMODE, freq=903900000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=904100000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=904300000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=904500000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=904700000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=904900000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=905100000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=905300000, len=26, SF=7, BW=125, CR=4/5, IH=0
TXMODE, freq=903900000, len=26, SF=7, BW=125, CR=4/5, IH=0
etc. etc. etc.
Is there anything that you can pick up from this @cstratton ?
You have 8 packets a minute from a single node?
What happens if you randomize the frequency rather than using them in turn? What if you increase the interval?
Not eight packets a minute from a single node. One packet roughly every minute from a single node. It doesn’t consistently stop at eight any longer. Sometimes it does 10. When it does 10, it frequency hops from 503 to 501 back to 503, and then like usual the node claims to send packets once a minute for 8-10 minutes before finally working again. The debug output always shows that frequency range previously named though. I haven’t tried increasing the interval. The payload is a uint8_t array, between 17-21 bytes.
I’ve got a project going on at my University deploying 5 sensors to measure air quality, so I’m hoping I’ll be able to finish setting these up before the Fall. On the bright-side, I kept working on everything else, and this is one of the last things I need to solve before the device should be a good working example for Github using the ESP32, RAK831 Pilot Gateway, PMS5003, BME280 and LoraServer. Optimizing the nodes for distance will be my next big learning curve, but first I need to attain consistency.