Video recordings from the 2017 Joint IEEE SPS and EURASIP Summer School on Signal Processing for 5G Wireless Access are available for IEEE members, as we wrote about in a previous post. Now two of the Massive MIMO tutorial talks are openly available on Youtube.
Prof. Erik. G. Larsson gave a 2.5 hour tutorial on the fundamentals of Massive MIMO, which is highly recommended for anyone learning this topic. You can then follow up by reading his book with the same topic.
When you have viewed Erik’s introduction, you can learn more about the state-of-the-art signal processing schemes for Massive MIMO from another talk at the summer school. Dr. Emil Björnson gave a 3 hour tutorial on this topic:
Regarding to the uncorrelated Rayleigh fading channel model, I know that the large-scale fading beta_ijk can be modeled as beta_ijk = z_ijk/(r_ijk/R)^alpha. I have 2 questions:
1) Does it mean that the channel quality does not depend on distance from user to BS but it depends on z_ijk: log-normal variable? So what does z_ijk mean in this model (scrattering, obstacle…).
2) How can I generate z_ijk with standard deviation = 8 dB on Matlab?
Thank you very much for your time!
1) With that model, the channel quality depends on the distance r_ijk
2) z = db2pow(randn(1)*8);
Is it fair to say that M-MIMO link budgets to a large extent depend on the channel assumptions and precoding scheme in use?
For example if you take a hypothetical LoS scenario, then beamsteering towards the AoA or AoD is probably optimal (I guess!), giving an array gain – no dependence on channel here, just on phase pre-compensation at the antenna array. But in an nLoS multipath environment is there a way to derive ZF or MRT link budgets for a generic channel. I think no (we’d always need multipath assumptions in place), but any comment is appreciated.
You put the finger on an important issue here: How do we measure SNR in a multiantenna system? Should we measure the SNR per antenna before we apply precoding or the effective SNR that we achieve after precoding? There are no simple answers to this, but I usually do the former to allow for comparison between different precoding schemes with fixed per-antenna SNR.
Thanks to the channel hardening and favorable propagation properties, the end performance doesn’t change much between different channel models. I would therefore recommend you to use SINR formulas for i.i.d. Rayleigh fading channels (even if such models are not realistic) in order to quantify the link budgets for a generic channel. You can find such formulas in “Fundamentals of Massive MIMO” or this book chapter: http://www.diva-portal.org/smash/get/diva2:1049059/FULLTEXT02.pdf