Collision between LoRa packets with different signal strengths


(Weffel) #1

Hello,

Does anyone know, or has sources of, what happens when 2 LoRa messages mix up?
Usually i see people state that if there are 2 incoming messages at the same time that overlap, both messages are gone. (Messages on same frequency and same SF)

But what happens when there is a weaker signal, at some point it is still strong enough to be received, but not strong enough to interfere with a far more powerful signal from a closer node. Is this possible?

To what difference in signal strength will it: have no effect, be filtered out, or corrupt both messages?


(Mirtcho) #2

I think the good starting point is this document
http://www.semtech.com/images/datasheet/an1200.22.pdf

In short Lora uses spread spectrum Chirp modulation . The chanse for collision is very small


(Arjan) #3

Because of the maximum duty cycle (or listen-before-talk in some regions), frequency hopping and use of different spreading factors? Or does the modulation itself even allow for multiple nodes transmitting on the same frequency and spreading factor simultaneously?


(Weffel) #4

Hi Mirtcho,

I have read that document before, I read it again, but I can not seem to find any specification of the results of a collision between 2 LoRa messages on the same frequency and same SF, with a different signal strength.

All I find is comparison and background information of other modulation techniques in respect to the LoRa modulation technique. And how they are filtered out because they are not similar to the LoRa modulation technique.


#5

Strongest packet is demodulated if power difference >6dB.


(Mirtcho) #6

Hi Weffel,
I don’'t kmow any beter document That explain LoRa spread spectrum.

I will try to oversimpliphy thinks. Later on you can re-read the simtech document
spread spectrum wiki

  • Resistance to jamming (interference
  • Resistance to eavesdropping
  • Resistance to fading
  • Multiple access capability - CDMA, WIFI and GPS are the most popular spread spectrum techniques.

Or over over simplified - Your and your neighbour WIFI can work at the same frequency channel. Yes both will work sub-optimal but none will be jammed. Even more the signal strenght of neighbour channel can be higher that your signal and you wil still receive data

@Hobo - Your statement is not valid for SpreadSpectrum. 6db is only valid for Narrow Band FSK or OOK modulation


#7

That’s what Semtech says.


(Jinoramson) #8

Hi,

This link is not working. Could you send me the document to ..@gmail.com


#9

(took me 2 seconds to find)


(Arjan) #10

Another nice resource as mentioned in another topic:


(Daryooosh) #11

Hi
So if two modules start to send different massage with same length at same time exactly,Which massage is received in receiver module?


(LoRaTracker) #12

From @arjanvanb post just above;

‘strongest packet is demodulated if power difference >6dB’


(Daryooosh) #13

Thanks
So if they have same strong…?
Another question
If two modules with same Sf,send two packet in small different time.what be happen(packet’s strong is same)?


(Ud Lo Ra) #14

The answer is here (0<6):


(LoRaTracker) #15

I am not sure if there has been an analysis by Semtech as to what happens when the time differences are ‘small’. It depends on how small is small, and the relation to the inevitable small difference in absolute frequency between nodes and drift during packet transmission.

Some collisions and conflict are inevitable, so why would it be important to know exaclty what happens in the range of possibilities ?