Why are the names of the 5G mmWave module using LTE and eNB

[chang yi Liu]

Hello everyone, I am trying to use NS-3 to simulate a 5G environment. I found the mmWave module, but after downloading it, I noticed that the representations inside use LTE and eNB. So, when I run the simulation, is it a simulation of 5G or 4G?  

[Jorge Ignacio Sandoval]

Hi,

you can try with 5G-LENA ( https://gitlab.com/cttc-lena/nr )

[Kent Huns]

It depends on what you recognize as the difference between 5G and 4G, 

but mmWave module from NYU and 5G-LENA from CTTC are both the packages to override MAC & PHY layer of LTE.(See Fig.1)

https://ieeexplore.ieee.org/document/8344116

As for 5G-LENA, the sources above the RLC layer have been distributed just recently, but I have heard that they are still just copies.

In any case, please read "README".

https://github.com/nyuwireless-unipd/ns3-mmwave

[chang yi Liu]

Thank you for the response.

I am a student still trying to understand the differences between 4G and 5G networks. So, is it possible to use mmWave for 5G NR? I am attempting to use mmWave because the handover feature in 5G-LENA has not yet been implemented. Or are there other modules in NS-3 that can be used for simulation?

Kent Huns 在 2024年11月2日 星期六晚上10:31:51 [UTC+8] 的信中寫道：

[Tommaso Pecorella]

@jordan - if you're trying to make sense of the 5G / 4G terminology, remember that the boundaries between "generations" are flexible, and in some cases quite confusing.

The standardization doesn't mandate "hard" changes between 3, 4, and 5G, and aims at implementing a gradual and smooth transition between each of them, minimizing the costs for the operators. Hence, one should for real refer to the "Releases" (see https://www.3gpp.org/specifications-technologies/releases). As an example, LTE has been standardized through Rel. 8 to Rel. 14 (see https://www.cablefree.net/wirelesstechnology/4glte/overview-of-lte-3gpp-releases/).

Note that the above does create a lot of confusion in the market, as one might claim that they have a "5G" network because they use something in Rel. 15+, even if they're missing almost everything. Plus, technically, LTE isn't 4G, because the "full" 4G is LTE-A... does it sounds crazy? Yes, it is. Standardization + marketing is an evil combo.

[Kent Huns]

>>is it possible to use mmWave for 5G NR?

Partly yes, mmWave & 5G-LENA support 5G's frequency bands, numerologies and beamforming algorithms. They enable us to simulate high throughput link(so-called eMBB).

And partly no, SDAP layer for network slicing and CU/DU/RU split for inter-cell coordination haven't been implemented.

   * (Correct me if I'm wrong)

      mmWave may support DualConnectivity, but I think it's similar to 5G-LENA in that it doesn't support disconnection(RLF) and initial access.
      https://groups.google.com/g/5g-lena-users/c/bxUu-MMhdII/m/ejc21n3RBAAJ 

      This means that mmWave doesn't support DC handover as well as you might imagine.

>>Or are there other modules in NS-3 that can be used for simulation?

Sorry for ambiguous answer, but I recommend you to figure out what you'd like to do at first.

There are so many type of "5G's handover", and the strict network simulator named ns-3 cannot ignore or abstract these differences.

You can see if you search this forum that many people have requested this feature and no one's wish has been come true.

I think it would be a good experience just to understand how difficult this modification is. Now you're gazing into the abyss.

If there is a simulator which supports (any type of)soft-handover, I guess quite a few researchers will be handed over to there.

https://www.youtube.com/watch?v=6dUtGjovlOo 

https://www.techplayon.com/5g-sa-handover-inter-gnb-du-and-intra-gnb-cu-handover/
https://telcomaglobal.com/p/different-types-handover-procedures-5g-connectivity