REG : [LBT Wi-Fi Coexistence] Same Node Consider for Calculation of pathloss

[saumil shah]

Dear All,

I am studying LTE/WiFi coexistence and using LBT Wifi coexistence module in ns3.

As we know based on prorogation model specified, loss between nodes will be calculated. Let's say we are using ItuInhPropagationLossModel model only which is given as default in indoor scenario.

As i think LOS or NLOS pathloss should be calculated between two different nodes and not the same node. For example in particular scenario there are two nodes x and y. So when pathloss will be calculated it should be between x - y and between y - x and not between  x-x and y-y.

But i have observe that whenever i use cellconfigA or cellconfigB as LAA cell or both as LAA cell then pathloss model is calculating pathloss between same node. But when i use cellconfigA and cellconfigB both as WIFI or LTE then this case is not there just when any of cell is LAA then only.

Reproduce the scenario:
1) I have modified  numUePerCell from 5 to 1 in "laa-wifi-indoor.cc" file

uint32_t numUePerCell = 1;

2) Modification in "itu-inh-propagation-loss-model.cc" file in the directory src/propagation/model.

in GetLosPathLossDb() and GetNlosPathLossDb() i have added below print statement after calculating distance between node a and b.

GetLosPathLossDb()
{
------
double d = a->GetDistanceFrom (b); // 3D distance
  std::cout << "distance = " << d << "  Node a = " << a->GetPosition() << "  Node b = " << b->GetPosition() << std::endl;
---------
}

GetNlosPathLossDb()
{
------
double d = a->GetDistanceFrom (b); // 3D distance
  std::cout << "distance = " << d << "  Node a = " << a->GetPosition() << "  Node b = " << b->GetPosition() << std::endl;
---------
}

After above modification i am running the script as below.

./waf --run "laa-wifi-indoor --cellConfigA=Wifi --cellConfigB=Wifi" >> output_wifi_wifi.txt
./waf --run "laa-wifi-indoor --cellConfigA=Wifi --cellConfigB=Laa" >> output_wifi_Laa.txt
./waf --run "laa-wifi-indoor --cellConfigA=Lte --cellConfigB=Lte" >> output_Lte_Lte.txt
./waf --run "laa-wifi-indoor --cellConfigA=Laa --cellConfigB=Laa" >> output_Laa_Laa.txt

If you check the generated .txt files you will find below whenever any cell is LAA.

distance = 0  Node a = 25:25:0  Node b = 25:25:0
distance = 0  Node a = 50:25:0  Node b = 50:25:0

Kindly check the generated file for more information or attached zip file.

Questions :

1) Is this case normal when pathloss will be calculated between same node?

2) Why is it happening like this for LAA cell only and not for WIFI or LTE cell?

3) If this is normal then while using LAA cell and suppose for LOSpath loss on given propagation loss model based on given logic distance 0 will be replaced with distance 3 as minimum model is defined for distance greater than 3 and pathloss will be calculated for distance = 3m. that means from same node to node we have pathloss which i did not find logical. Can anyone explain this?

4) as explained in question 3 we will get path loss between same node. Will this value will be consider for any further purpose?


PS : I was implementing my own pathloss model and it has one term log(distance). Unfortunately i have not handled case when distance = 0 as i have not thought that pathloss will be calculated between same node. So whenever distance was 0 log(0) was giving infinity and other cell was not able to start only..i was just getting output for Laa only and not wifi. So when i was debugging i found this problem and i have tried with already implemented path loss model and observed the same thing. Anyway later i have handle the condition for distance = 0 but still it is strange why it is happening like this.

Best Regards
Saumil Shah

[Rediet]

Hello,

Have a look at https://groups.google.com/forum/#!topic/ns-3-users/Ye3poGpRkno and https://groups.google.com/forum/#!topic/ns-3-users/r0le2dBZT1w.

Cheers,

Rediet

[saumil shah]

Hello Rediet,

Thanks for your suggestion. I have checked the both the discussion and i have also handled the same way as for distance 0.

But do you have idea about question 1 and question 2 in my initial post?

Best Regards,
Saumil

[Rediet]

Hey Saumil,

That was I redirected you to the second post. In order to implement LBT feature for LAA, the project team chose to model this using the existing Wi-Fi LBT feature; a Wi-Fi ad-hoc network device was added to the LTE eNB so that the LTE network device may be able to listen to what's going on on the unlicensed channel before transmitting (i.e. over the DL supplemental channel). Therefore, there are 2 NetDevice working on the same unlicensed channel for a LAA eNB (for Rel 13); the LTE for transmitting and the Wi-Fi ad-hoc for receiving. That's why you see pathloss calculated between the two devices.

Rediet

[saumil shah]

Hey Rediet,

Now i get it and was able to relate that second post with my question. Thanks for the explanation.

So basically loss will be calculated between two different NetDevices and when both NetDevices are installed on the same node we should handle that condition with distance = 0 and returning the loss as 0.

Best Regard
Saumil