Vraag

The Behavior Difference of Lora Mote

  • 27 juli 2015
  • 5 reacties
  • 1002 keer bekeken

Hello,

I am an student doing research on Lora Mote.

I have also get a lora Gateway device which is indoor currently.

I first program the Mote with the GUI of "LoraMacConfig". The Lora device is found by KPN wlogger.

The spreading factor is 9.

Afterwards, I use the program "LoRa_Keil5LoRaMac-node-masterKeilLoRaMoteLoRaMacclassA"

 downloaded fromthe website GitHub "LoRa™ network". The behavior shows some difference.

The LoraMote works not as stable as prorammed with LoraMacCofig. The spreading factor drops mostly to 12 and some times recovered up to 7.

In the attached the photo show the behaviour difference.

Anyboday has some idea how this difference is caused? Is there a way to improve the Mote behavior with the classA code?

Thank You!

5 reacties

Reputatie 1
Badge
Below is our frequency channel setup:

Channel:0 Not Used

Channel:1 LC1, Subband=1, freqhz=868.10 (125KHZ, DR_LORA_MULTI)
Channel:2 LC2, Subband=1, freqhz=868.30 (125KHZ, DR_LORA_MULTI)
channel:3 LC3, Subband=1, freqhz=868.50 (125KHZ, DR_LORA_MULTI)
 

Used for rx2 = 1 Data rate rx2= DR_LORA_SF9 power=+0
Reputatie 1
Badge
The difference between these could be caused in frequency channels used and the way the application has been setup to use confirmed and unconfirmed messages. Can you set your used channel mask to 0x0007 and try again.
Reputatie 1
Badge
At the moment the KPN developer gateway uses 6  channels on which only the first 3 are LoraWAN specified meaning these are the same over Semtech (https://github.com/Lora-net/LoRaMac-node) or IBM (http://www.research.ibm.com/labs/zurich/ics/lrsc/lmic.html) code.

We are evaluating if 3 non LoraWAN specified channels need to be changed in our setup.

If you want to make sure the channels are supported you can apply the channel mask of 0x0007.
So, is the gateway only using the smallest set of channels (868.10, 868.30, 868.50)?
Thanks.

Are supported frequencies/datarates and other characteristics documented somewhere?

Reageer