RLF Detection in LTE Module

[Çağrı DEMİR]

Hello Experts, 

I am using NS3 version 3.36 and trying to simulate an environment with multi UE and multi eNBs to measure the handover statistics of the network. For this, I used the example script (lena-x2-handover-measures) and modified some parameters such as channel propagation model, inter-cell distance, enb Tx power etc. 

Basically, eNB Tx Power: 46 dBm, inter-cell distance = 500 meter and pathloss model set as below in my scenario.

lteHelper->SetPathlossModelType (TypeId::LookupByName ("ns3::LogDistancePropagationLossModel"));
  lteHelper->SetPathlossModelAttribute ("Exponent", DoubleValue (3.5));
  lteHelper->SetPathlossModelAttribute ("ReferenceLoss", DoubleValue (38.57)); 
  lteHelper->SetPathlossModelAttribute ("ReferenceDistance", DoubleValue (1));

Before configuring the advanced system parameters (defining multiple UEs and eNBs with different config. parameters), I decided to find out the borders of the cells when specific Tx Power, distance and chanel model used. By border of cell, I mean: in what RSRP/SINR UE detecting RLF. I will use this while setting handover parameters between eNBs. (Hyst, TTT etc.)

To achieve this, first, I have set only one eNB and one UE and define mobility for the UE from the center of cell through the outside of the cell. Based on the data, RLF detection is happening at around SINR == -5 dB (I am checking SINR values in the log file named: DlRsrpSinrStats.txt). I think this makes sense because "m_qOut" attribute set as -5 in lte-ue-phy.cc file.

However, when I enabled 2nd eNB and set the inter-cell-distance between eNBs as 500m, with exact same mobility scenario, RLF is detected by UE at around SINR == 13 dB. After RLF detection, UE tries to attach second eNB as It has more RSRP than the first one, as expected. 

I am trying to understand how does the RLF detection triggered in the LENA module? I have checked the implementation in lte-ue-phy.cc and found out that SINR is calculating for last 10 frame and SINR for each frame compared with "m_qOut" attribute. If the calculated SINR is less than "m_qOut"  for 20 frames consecutively (m_numOfQoutEvalSf), RLF is being triggered. 

But why in DlRsrpSinrStats.txt file, I am seeing SINR == 13 dB when the time RLF detected with 2eNBs? I would expect to see SINR == -5 dB similar to the scenario with 1 eNB. Please help me to understand.

Thanks in advance for your comments,

Cagri

[Çağrı DEMİR]

Hello, 

After checking the code in more detail, I have realized that the RLF detection performed based on the channel quality of PDCCH (control channel). But the RSRP/SINR values in DlRsrpSinrStats.txt belong to the PDSCH (data channel) data. So it might be normal to have an RLF with high SINR in PDSCH. Because the PDCCH is still below the threshold. (-5 dB)

However, about the difference between 1 eNB and 2 eNB scenarios, I have seen neighbor eNB only interfere with the PDCCH channel but not PDSCH. Please have a look at the below graphs. 

SINR value for PDCCH during the simulation is less and degrades faster when 2 eNB is defined in the simulation scenario.

But the behavior is not the same for PDSCH. Seems like there is no interference effect applied to the PDSCH no matter how many eNB defined in the simulation scenario. PDSCH charts exactly the same in both 1 and 2 enb scenarios.

Is this normal behavior? I would expect to have interference also in the data channel (PDSCH) as well when the numOf eNB increased.

Note: Test performed with 1 UE and is both UL/DL throughput already defined in the simulation script. Below is the tbSize graph for the UE in DL in 1 eNB simulation.

Thanks,

Cagri

Çağrı DEMİR <cagri....@gmail.com>, 29 Kas 2022 Sal, 01:20 tarihinde şunu yazdı:

--
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/WhYHwRSj41Q/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ns-3-users/17675781-554c-4d7e-adc0-fc095b39407dn%40googlegroups.com.