search

nyuwireless-unipd / ns3-mmwave

ns-3 module for simulating mmWave-based cellular systems. See https://ieeexplore.ieee.org/document/8344116/ (open access) as a reference.
GNU General Public License v2.0
298 stars 193 forks source link

Mmwave-simple-building-obstacle UL-DL mcs value problem #86

Closed Faydin94 closed 5 years ago

Faydin94 commented 5 years ago

Hi, I am working on mmwave-simple-building-obstacle example. For different UE velocity values I ran the example. For all cases UL mcs and hence tbsize values are greater than those of DL even the SINR values are the same (as you can see on figure). What is the reason for this? If you are using a table to assign mcs values for SINR ranges, where can I find it? Regards, Furkan

simple building obstacle samples

tommasozugno commented 5 years ago

Hi, This is weird, I tried with other examples but I did not find any relevant difference. What is the speed of the UE you are considering in this case?

The MCS is computed in the scheduler using the methods GetMcsFromCqi () and CreateCqiFeedbackWbTdma () of the MmWaveAmc class. For instance, in MmWaveFlexTtiMacScheduler is computed here for the DL https://github.com/nyuwireless-unipd/ns3-mmwave/blob/7c25c1b91d005374746c1529a3bed5beb849a455/src/mmwave/model/mmwave-flex-tti-mac-scheduler.cc#L1008 and here for the UL https://github.com/nyuwireless-unipd/ns3-mmwave/blob/7c25c1b91d005374746c1529a3bed5beb849a455/src/mmwave/model/mmwave-flex-tti-mac-scheduler.cc#L1084

Best, Tommaso

Faydin94 commented 5 years ago

I ran the example firstly with default velocity values and then with (0,30,0),(0,10,0),(0,0,0) values. The results are as I mentioned before. Now, I build a new-handover branch from scratch. After ran the example, I saw that for UL: SINR=2.90 dB mcs=9 tbs=3617
for DL: SINR=3.87 dB mcs=8 and tbs=3118 Could it be a mapping problem only for this example for DL or UL? Best, Furkan

tommasozugno commented 5 years ago

Hi Furkan,

This is because the average SINR value that you see in the trace is the one measured at the time when the TB transmission takes place, while the MCS value is the one that has been computed by the scheduler when the transmission was scheduled.

Best, Tommaso

Faydin94 commented 5 years ago

Hı, I see your point, thanks. Then, how can I trace SINR for comparison of DL and UL cases in synchronized manner. Does any shifting operation in time does handle the problem? Best, Furkan

tommasozugno commented 5 years ago

Hi,

You can use the LOG messages of the MmWaveAmc or the MmWaveFlexTtiMacScheduler classes.

Best, Tommaso

Faydin94 commented 5 years ago

Thanks