Regarding the working of listen-before-talk

shobhuagg · May 26, 2020, 5:04pm

Hello everyone,
I some confusion regarding LBT mode in LoRaWAN.
So, in LBT mode the end device opens a receive window before transmitting. I am confused that what is the configuration of that receive window.

In other words, during the receive window the end device listen to the entire channel (all spreading factors) or a specifically configured transmission on that channel?

Thanks,

jreiss · May 27, 2020, 2:40pm

LBT should be modulation/datarate agnostic and simply read the RSSI for the selected channel before transmitting.

An example implementation can be seen in the isChannelFree function for the Sx1272 driver.

github.com

Lora-net/LoRaMac-node/blob/v4.4.3/src/radio/sx1272/sx1272.c#L257-L288


bool SX1272IsChannelFree( RadioModems_t modem, uint32_t freq, int16_t rssiThresh, uint32_t maxCarrierSenseTime )

{

    bool status = true;

    int16_t rssi = 0;

    uint32_t carrierSenseTime = 0;



    SX1272SetSleep( );



    SX1272SetModem( modem );



    SX1272SetChannel( freq );



    SX1272SetOpMode( RF_OPMODE_RECEIVER );



    DelayMs( 1 );



    carrierSenseTime = TimerGetCurrentTime( );



    // Perform carrier sense for maxCarrierSenseTime

    while( TimerGetElapsedTime( carrierSenseTime ) < maxCarrierSenseTime )

This file has been truncated. show original

shobhuagg · May 29, 2020, 11:50pm

Thanks for your response but why should LBT be datarate agnostic?

Suppose one of the end device transmits on channel-A SF7 and at the same time other device wants to transmit on channel-A but at SF-10, according to LBT the channel is busy, but practically the message can be sent successfully.

Am I missing something here?

cslorabox · May 30, 2020, 1:51am

The other use of the channel in progress might not even be LoRa

While it’s true that there is a limited degree of orthogonality of certain spreading factor / bandwidth combinations when the signal levels aren’t drastically different, that only protects the LoRa side. Another channel user might well be using a modulation which would be more readily interfered with by a LoRa transmission.

Essentially the answer to both questions is “because the spectrum is not reserved for LoRa alone”