REG : [LAA/WIFI Coexistence] indoor scenario error msg = "No frame sync error rate data stored for snr key = -2147483648"

[saumil shah]

Dear all,

I am running simulations for indoor scenario in LAA/WIFI coexistence module.

And i am getting below error:

msg="No frame sync error rate data stored for snr key = -2147483648", file=../src/wifi/model/frame-sync-error-rate-lookup.cc, line=421 terminate called without an active exception

script : ./waf --run "ss_laa_wifi_indoor --duration=15 --num_AP_OP_B=1 --num_AP_OP_A=10"

Description:

I have pre-generated location of all the nodes and stored in .csv file. I have modified the code in a way that it will read node locations from file based on number of nodes for Operator A and Operator B which i will provide with  --num_AP_OP_B=1 --num_AP_OP_A=10.

Here Operator A is LAA with DutyCycle option which has 10 pairs of eNB and UE , Operator B is Wifi which has 1 pair of AP-STA.

I have sent RngRun = 21245.

Steps to Reproduce Error:
1) Put attached bs_A_labels_21245.csv,bs_B_labels_21245.csv,ue_A_labels_21245.csv,ue_A_labels_21245.csv files into /ns-3-lbt directory.
2) Put attached ss_laa_wifi_indoor.cc file into /ns-3-lbt/src/laa-wifi-coexistence/examples directory and add this file into wscript too.
3) Put attached scenario-helper.cc file into /ns-3-lbt/src/laa-wifi-coexistence/helper directory.
4) Run the simulation with script : ./waf --run "ss_laa_wifi_indoor --duration=15 --num_AP_OP_B=1 --num_AP_OP_A=10"

Some Notes:
1) This error is coming for only some combinations of num_AP_OP_B and num_AP_OP_A.
for example, --num_AP_OP_B=1 --num_AP_OP_A=9 , --num_AP_OP_B=5 --num_AP_OP_A=10 ,  --num_AP_OP_B=1 --num_AP_OP_A=10 for these combinations i am getting this error

2) For many other combinations i am not getting this error.
for example, --num_AP_OP_B=5 --num_AP_OP_A=3  ,  --num_AP_OP_B=5 --num_AP_OP_A=5 ,  --num_AP_OP_B=5 --num_AP_OP_A=6 and many others i am not getting this error.

Questions:
1) Can someone explain me what does this error means and how should i correct it?
2) Why is this happening for some combinations and not for some?

Best Regards
Saumil Shah

[saumil shah]

Dear All,

In continuation to the above post i am able to track down the problem till one level as explained below which might be helpful to debug this error.

1) This error occurred during the function call InterpolateAndRetrieveData (double xd, double x1d, double x2d) which is defined in /ns3_lbt/src/wifi/model/freq-selective-error-rate-model.cc file.
I have tried to print the value of xd, x1d and x2d which is nan for all three.
xd  = nan
x1d = nan
x2d = nan

2) Now this function is called from the function GetFrameSyncErrorRate (double snr) which is defined in /ns3_lbt/src/wifi/model/freq-selective-error-rate-model.cc file.
When i try to print the value of SNR in the beginning of this function it also prints nan .

3) Now function GetFrameSyncErrorRate (double snr) is called from function SpectrumWifiPhy::StartRx (Ptr<SpectrumSignalParameters> rxParams) which is defined in /ns3_lbt/src/wifi/model/spectrum-wifi-phy.cc file.

if (m_random->GetValue () > m_frameSyncErrorRateModel->GetFrameSyncErrorRate (RatioToDb (m_interference.CalculateAvgSyncFieldSnir (event))))

4) Now when i traced the function CalculateAvgSyncFieldSnir() which is defined in /ns3_lbt/src/wifi/model/interference-helper.cc file it calls another function InterferenceHelper::CalculateSnr() which is also defined in same file.

In this function i have printed signal power , noise power, noise floor, interference , BW and SNR (linear) which gives me below output when error occurs.

bandwidth(MHz)= 20, signal(W)= 2.66269e-12, noise(W)= 0.199526, noisefloor(W)= 6.35919e-13, interference(W)= 0.199526, snr(linear)= 1.33451e-11
bandwidth(MHz)= 20, signal(W)= 1.8655e-09, noise(W)= 0.199526, noisefloor(W)= 6.35919e-13, interference(W)= 0.199526, snr(linear)= 9.34963e-09
bandwidth(MHz)= 20, signal(W)= 1.8655e-09, noise(W)= 0.199526, noisefloor(W)= 6.35919e-13, interference(W)= 0.199526, snr(linear)= 9.34963e-09
bandwidth(MHz)= 20, signal(W)= 9.96694e-08, noise(W)= 3.05965e-09, noisefloor(W)= 4.01237e-13, interference(W)= 3.05925e-09, snr(linear)= 32.5754
bandwidth(MHz)= 20, signal(W)= 9.96694e-08, noise(W)= 3.05965e-09, noisefloor(W)= 4.01237e-13, interference(W)= 3.05925e-09, snr(linear)= 32.5754
bandwidth(MHz)= 20, signal(W)= 1.60286e-12, noise(W)= -2.39371e-15, noisefloor(W)= 6.35919e-13, interference(W)= -6.38312e-13, snr(linear)= -669.612

now as marked in last line we are getting noise as negative value so SNR will also be negative.

5) Now when this SNR will be pass to RatioToDb() function to convert into log scale ,it will give value as nan because we are trying to take log of negative value which is not defined.

Now i don't know further why we are getting as noise negative and what can be the reason behind this.
Kindly help me with this issue.

Thanks
Saumil

[pdbarnes]

Thank you Saumil for tracking this down as far as you did.

I've filed bug 2846 to track it

https://www.nsnam.org/bugzilla/show_bug.cgi?id=2846

[saumil shah]

Dear Peter,

Thank you for looking into this issue.

Kindly let me know if need to test some solution or patch for this issue. I will be happy to help.

Best Regards
Saumil

[pdbarnes]

The big question is if you can reproduce the error. If so we can suggest a patch to help diagnose the problem, as I suggested in the bug report.

Peter

[saumil shah]

Dear Peter,

I believe i can reproduce the error. I have got some 4-5 different RngRun value for which i have observed this error and i have my scripts, so i think i can reproduce it.

Best Regards

Saumil

On 9 Jan 2018 8:10 pm, "pdbarnes" <pd...@mac.com> wrote:

The big question is if you can reproduce the error. If so we can suggest a patch to help diagnose the problem, as I suggested in the bug report.

Peter

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/t6L35V_eZ8M/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+unsubscribe@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

[Sebastien Deronne]

Can you please provide a script that reproduces the issue in ns-3-dev?

Le mardi 9 janvier 2018 22:24:50 UTC+1, saumil shah a écrit :

Dear Peter,

I believe i can reproduce the error. I have got some 4-5 different RngRun value for which i have observed this error and i have my scripts, so i think i can reproduce it.

Best Regards

Saumil

On 9 Jan 2018 8:10 pm, "pdbarnes" <pd...@mac.com> wrote:

The big question is if you can reproduce the error. If so we can suggest a patch to help diagnose the problem, as I suggested in the bug report.

Peter

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/t6L35V_eZ8M/unsubscribe.

To unsubscribe from this group and all its topics, send an email to ns-3-users+...@googlegroups.com.

[saumil shah]

Dear Sebastien,

In the first of post of this thread i have attached all the necessary files and script. Do you need something else or that's what are you asking?

script : ./waf --run "ss_laa_wifi_indoor --duration=15 --num_AP_OP_B=1 --num_AP_OP_A=10"

Best Regards

Saumil

[Tom Henderson]

--

Hi Saumil,
I was able to reproduce the problem and it seems to be due to the bug 2845: https://www.nsnam.org/bugzilla/show_bug.cgi?id=2845

The below patch cleared the issue for me for the LAA code:

diff -r fb0495e70736 src/wifi/model/interference-helper.cc
--- a/src/wifi/model/interference-helper.cc     Sat Dec 09 11:46:26 2017 -0800
+++ b/src/wifi/model/interference-helper.cc     Wed Jan 10 19:36:22 2018 -0800
@@ -284,6 +284,10 @@
     }
   ni->insert (ni->begin (), NiChange (event->GetStartTime (), noiseInterference));
   ni->push_back (NiChange (event->GetEndTime (), 0));
+  if (noiseInterference < 0)
+    {
+      noiseInterference = 0;
+    }
   return noiseInterference;
 }

@@ -900,6 +904,10 @@
     {
       averageInterferencePowerW = weightedTotal / sumOfWeights; // weighted average
     }
+  if (averageInterferencePowerW < 0)
+    {
+      averageInterferencePowerW = 0;
+    }
   return CalculateSnr (powerW, averageInterferencePowerW, txVector.GetChannelWidth ());
 }

[saumil shah]

Dear Tom,

Above patch also solved problem for me. Thank you for your help.

Will it be the final solution for the bug or is it a temporary patch?

Best Regards
Saumil

[Tom Henderson]

On 01/12/2018 05:57 AM, saumil shah wrote:
> Dear Tom,
>
> Above patch also solved problem for me. Thank you for your help.
>
> Will it be the final solution for the bug or is it a temporary patch?

The final solution is being worked on (see bug 2791 in our tracker).

- Tom

[pdbarnes]

Hello Saumil,

Sebastien recently pushed a fix to ns-3-dev.  Could you please try your problem again and report if it still persists?

We are tracking this in this bug report:

https://www.nsnam.org/bugzilla/show_bug.cgi?id=2791

Thanks,

Peter