What is Multi-User MIMO Communications (MU MIMO)?

Explains the main points of multiple user multiple input multiple output (MU MIMO) digital communications.
Related videos: (see: iaincollings.com)
• MIMO Communications • MIMO Communications
• What is Massive MIMO? • What is Massive MIMO?
• What is Beamforming? • What is Beamforming? (...
• What is Multi Channel Beamforming? • What is Multi Channel ...
• How are Beamforming and Precoding Related? • How are Beamforming an...
• Quick Introduction to MIMO Channel Estimation • Quick Introduction to ...
• Why doesn't MIMO work in Line-of-Sight (LoS) Channel Conditions? • Why doesn't MIMO work ...
• How does Antenna Spacing affect Beamforming? • How does Antenna Spaci...
• OFDM Waveforms • OFDM Waveforms
• OFDM and the DFT • OFDM and the DFT
• What is a Cyclic Prefix in OFDM? • What is a Cyclic Prefi...
• What is Rayleigh Fading? • What is Rayleigh Fading?
• What is Rician Fading? • What is Rician Fading?
• What are Fast Fading and Slow Fading? • What are Fast Fading a...
• What are Flat Fading and Frequency Selective Fading? • What are Flat Fading a...
Full categorised list of videos and PDF Summary Sheets: iaincollings.com
.

[@j.m.2143]

MU-MIMO and CSMA / CA Hello everybody, I am trying to understand the concept of MU-MIMO introduced in 802.11ac. Supposedly the WiFi AP can send multiple data frames to multiple users at the same time, how is it possible to do this without breaking the CSMA / CA standard ??? I understand that if we have a 20 MHZ channel and 8 users, the 8 will launch their RTS until the AP gives their CTS to transmit, it is like a round table where one "speaks" and the rest listens because all the users are competing channel and only one can send / receive data at the same time. On the other hand, in MU-MIMO, several users can receive data on the same channel at the same time, WHICH DOES NOT MAKE SENSE. Can someone clarify which part I don't understand? Because if it were an OFDMA modulation would include assigning 36 carriers to one device and the other 24 to another, but in 802.11ac all carriers are assigned to the same device and to the entire channel.

[@jenniferandrea7996]

Prof, Thank you for great series of lectures. What do you mean by referring Narrow band representation to channel gain ( being complex value) at @6:40 ? Can channel gain be complex ? what does imaginary value indicates ? Appreciate your response! Thanks

[@iain_explains]

These videos will hopefully help: "How are Complex Baseband Digital Signals Transmitted?" kzbin.info/www/bejne/Zp3Og32do96qock and "What is a Baseband Equivalent Signal in Communications?" kzbin.info/www/bejne/m6W9coWXgrOBaNU

[@oldPrince22]

@@iain_explains Is "narrowband" and "baseband" meaning the same thing here?

[@dimitrisv.1729]

Concerning the parallel implementation for each subcarrier in OFDM. For example if you have 4 single-antenna users and the length of IFFT/FFT is 256, you mean that you have to decode 1x4 y vectors in parallel for 256 times? Effect of channel correlation in MIMO systems(and massive MIMO) would be interesting as a next topic.

[@iain_explains]

Yes, that's right. And I've added channel correlation to my "to do" list.

[@prabhatsahoo7535]

Its a very informative. If the sequence of the description can be as below, it will be more useful - A simple explanation about what is MU-MIMO - Mathematical explanation - Benefits of MU-MIMO

[@iain_explains]

Glad you liked it. And thanks for the suggestion. In this video, I was really only aiming to answer the question in the title, at a general level. The Maths and Benefits can come in another video. I'll put it on my "to do" list.

[@larrygilman2919]

At about 6:30 you remark that for the uplink, synching training is needed "if it's not time-division duplexing." Could you say a word about why this is the case? What here might be time-division duplexed (or not), and how would this bear on synching the training signals? Many thanks. Love your pedagogical combo of pen, paper, and speech.

[@iain_explains]

Ah, thanks, I see, I didn't explain that very well. If it is a TDD system, then you actually don't need to send training symbols on the downlink. In the TDD case, the same frequency band is being used on both uplink and downlink, and so you only need to send training on the uplink, measure the channel at the base station, and then use that as the channel estimate for the downlink, since there is "channel reciprocity". In FDD systems, all users need to know when the downlink training is going to be coming, and if they are at different distances from the base station, then it will arrive at the different users at different times, and they need to know that (this is what I was trying to say). And thanks for your comment about my pedagogical approach, I really appreciate it. I try to create a "personal tutoring" feeling, as if I'm sitting next to the viewer. I aim for content and understanding, rather than "graphics and entertainment". Glad you like it.

[@larrygilman2919]

@@iain_explains Thanks for the reply -- that really clears it up.

[@iain_explains]

My pleasure.

[@sivasaikrishnamarthy178]

Dear Professor, At @3.43 How come rich scattering in MIMO is related to H inverse here, ? Why does H inverse is not possible during line of sight scenarios ?

[@iain_explains]

Here's a video on that topic: "Why doesn't MIMO work in Line-of-Sight (LoS) Channel Conditions?" kzbin.info/www/bejne/rqLQlYKOnb9lepY

[@bobbaberson3654]

Maybe a basic question. Why is H is not invertible if there is LOS or if the scattering is not rich?

[@iain_explains]

Good question. If there was only a line of sight path, (and no scatterers providing alternate delayed paths with different departure and arrival angles), then the gain between each of the transmit antennas and each of the receive antennas would be (almost exactly) the same. This would mean that every row of the channel matrix would be the same. And in turn, this means that the matrix would not be able to be inverted (because there would only be one non-zero Eigen value). These videos discuss this more: "What are Spatial Diversity and Spatial Multiplexing in MIMO?" kzbin.info/www/bejne/g3-kYaukbKqYr7s

[@bobbaberson3654]

@@iain_explains Your ZF video made me think about this again :D... In a SU MIMO system, even rich multipath should not matter much right? As far as I understand, H is an nxm complex-valued matrix, if since the antennas are closely spaced at a user device, the matrix would still be non-invertible, is that right?

[@shahriarsowad9221]

Dr. Lain, I completed all your video lectures on MIMO. I am interested in exploring the field of Holographic MIMO, not finding any good resources to start my journey. Can you suggest me some?

[@iain_explains]

Thanks for the suggestion. I've put it on my "to do" list.

[@kindakhawam1575]

So elegant !

[@iain_explains]

Thanks. I'm glad you think so.

[@omar778]

Thanks

[@DVVSNReddy]

Prof make a playlist on mimo channel pre-coding & deep learning in wireless comms (in PHY).

[@iain_explains]

The following video discusses the basics of MIMO precoding: kzbin.info/www/bejne/inSUapqDhpuUsKc I'll give some thought to more on this topic, plus your deep learning suggestion. Thanks.

[@sandeepreddy8567]

If we have a data stream- 'abcd', do we send 'a', 'b', 'c', 'd' from 4 different antennas in gNodeB (increasing its throughout) to the UE in different paths? If true, y do we use CP-OFDM in 5G when we sending different data streams ! or do we send 'a' from all 4 antennas at one instance & 'b' from all 4 antennas at other instance & so on (so that UE can choose the best signal with higher power) ?

[@iain_explains]

The MIMO precoding at the transmitter is done (separately) on each OFDM subcarrier. So you actually have both spatial sub channels (using MIMO), and frequency sub channels (using OFDM), at the same time.

[@tadiosbelay2307]

dear prof , what is AOA and AOD in MU-MIMO

[@iain_explains]

There will be a different angle of arrival and angle of departure for each user, depending on their location, and the channel between them and the multi-antenna base station.

[@ravikiran412]

Sir can you explain which techniques we are using in MU MIMO? And suppose in transmitted side we have 2antennas with different stream each antenna have and we have 2users in the receiver side but one user have one antenna and other user have 2antennas.In this case how will you decode the transmitted signal?

[@iain_explains]

Thanks for your question, but there's not one clear answer, I'm sorry. There are quite a few different ways to implement MU-MIMO. It all comes down to how you choose the precoding matrix. I've put the topic on my "to do" list, but in the meantime you might like to watch this video (if you haven't already): "How are Beamforming and Precoding Related?" kzbin.info/www/bejne/n36sonimr92Bb7s

[@ravikiran412]

@@iain_explains y eah, I have seen this video.. Precoding means multiple beamforming.

[@yasserothman4023]

Thank you but I wanted to know 1- In siso systems assuming 64 ofdm symbol , is it the case that the 64 subcarruers are upconverted and sent all at once so that the receiver gets a 64 by one vector or is it that each subcarruer is up converted and sent so that we receive a scalar and after 64 time instances we start processing the subcarriers ? 2-How does it differ in case of mimo systems ? 3- I didn't get the notion of parallel implementation that you pointed out at the last part of the video ? And why is it specific only for mu-mimo Thanks

[@iain_explains]

For Q1, I suggest watching my video on OFDM. kzbin.info/www/bejne/kGWvepqEnLN0oqs It explains what happens at baseband. It doesn't explain up conversion in that video, but it's straightforward. The time-domain signal generated by the IDFT/IFFT is upconverted to the carrier frequency, simply by multiplying by the sinusoidal carrier. This has the effect of up converting all the subcarriers in the one operation. For Q2, the MIMO operation needs to be done separately on a subcarrier, by subcarrier, basis, before sending the resultant symbols to the OFDM module in the transmitter. The MIMO operation, as described in kzbin.info/www/bejne/inSUapqDhpuUsKc is single-carrier (or narrow band). For Q3, I hope the above two answers have explained this.

[@gopalsolanki1519]

@@iain_explains Professor, Can you please create a video that expands on your answer here. i.e. MU-MIMO + OFDM from one end to the other (including up conversion). Just so that everything comes together in one place. Will greatly appreciate it. It will also be great if you can add note about what is the simplified equalization that still needs to be done in the OFDM systems to tie the whole story of say LTE or 5G communication together in one place end to end.

[@safiyajd]

What is synchronization between users for? Thanks for the great content.

[@iain_explains]

In the uplink of a MU-MIMO system, the symbols from all the users need to arrive at the same time, so the users need to have accurate timing information (more accurate than what's needed for TDMA).

[@tianyuez]

Respect!

[@iain_explains]

Glad you are finding the videos useful.

[@j.m.2143]

Hello greetings. MU-MIMO no longer uses CSMA / CA, RTS, CTS?

[@iain_explains]

Yes, you make a good point. MU-MIMO on the uplink requires coordination among the users, in terms of packet transmission timing and allocation to slots. Also, the base station needs knowledge of all the users' channels if it wants to point the receive beams directly towards the users. Strictly speaking it's not suited to random access channel sharing schemes. However, some variations have been proposed, which are also sometimes just called MU-MIMO too, which are incorporated into some standards, eg. 802.11ax.

[@王大維-w4t]

Could you make a video about massive MIMO? I got to give a presentation next week but am a bit confused about the difference between MU MIMO and massive MIMO. Thanks!

[@iain_explains]

Thanks for the suggestion. This topic is on my "to do" list, but unfortunately I'm not going to be able to get to it before your presentation, sorry.

[@王大維-w4t]

@@iain_explains No problem! It would also be great if you could talk about the channel hardening effect.

[@iain_explains]

Thanks for the suggestion. I've added that to the list too.

[@issaclee7114]

Could you explain the difference between MU-MIMO and massive MIMO?

[@dimitrisv.1729]

Let N denote the number of antennas at base station and K the number of users. In massive MIMO systems, N>>K is assumed. This leads to some properties that are exploited in massive MIMO such as channel hardening.

[@Will-bx2wz]

Can you talk about the technique used when H is invertible but very rank deficient (high correlation between channels)?

[@iain_explains]

When there's high correlation between the channels, then it's a good idea to consider using beam forming. These videos discuss this more: "What are Spatial Diversity and Spatial Multiplexing in MIMO?" kzbin.info/www/bejne/g3-kYaukbKqYr7s and "What is Beamforming?" kzbin.info/www/bejne/d2LRZnuerNmtbZo

[@윤덕호-g9v]

done