How does Antenna Spacing affect Beamforming?
Вставка
- Опубліковано 10 вер 2020
- Demonstrates how the number of antennas and their spacing affects beam forming.
Related videos: (see: iaincollings.com)
• What is Beamforming? • What is Beamforming? (...
• How are Beamforming and Precoding Related? • How are Beamforming an...
• What is Hybrid Beamforming? • What is Hybrid Beamfor...
• What is Multi Channel Beamforming? • What is Multi Channel ...
• What is a Matched Filter? • What is a Matched Filter?
• How are Correlation and Convolution Related in Digital Communications? • How are Correlation an...
For the Matlab code, check out the full categorised listing of videos at: iaincollings.com
.
thank you
This is just great!
Thanks. Glad you found it useful.
Awesome sir
Thanks. I'm glad you liked it.
Thanks for all your informative videos. They helped me greatly. Could you kindly let me know where are the Matlab codes in your website. I couldn't find them .
It is under "Digital Communications" then "Multiple Antenna Communications". Anyway, here's the direct link: drive.google.com/file/d/1faZpLKNaElCqD7e1yZ5mj6SLwr9nWG37/view
@@iain_explains thank you so much
Thanks for your materials. I wonder why the representation is done for (-90,90) and not (-180,180) so that we cover the complete 360 degrees 🤔
It's a linear array, so there is a direct symmetry (reflected) on the other side. Although, in most practical cases, a backplane is installed behind the array, so that radiation only goes out on one side of the array (eg. to not waste energy radiating into a building that the array is attached to).
Prof, I've got a question, if the size of the array is fixed, and we add the number of antenna elements by decreasing the spacing of elements, and neglect the loss of radiating efficiency caused by active impedance matching, would the beam pattern be "narrower" and "sharper" like in the video you showed? Or in another word, would the characteristic of main lobe for a directivity pattern change by adding elements for a fixed-size array?
If you're needing help to visualise it, you can try it for yourself using the Matlab code that I supplied. You can find the link at my website: iaincollings.com (or here's the direct link: drive.google.com/file/d/1faZpLKNaElCqD7e1yZ5mj6SLwr9nWG37/view?usp=sharing )
@@iain_explains Thanx for answer. Not what i meant though. If planewave assumption, the distance d is wrongly assumed to be a hypothenuse, when it in a linear subspace in reality is the catheter of a triangle to a bigger circle BELOW the array baseline...? . So if the angle is tangent, not cosine. the scaled/rotated/subspace radius can be found using only three sensors at any mutual distance on a straight line without iterations. Arctan left, arctan right, will give the exact and TRUE angle, and the distance to array baseline from two pairs of sensors.where those normals cross will also be the exact radius at T0...The plane wave assumption is in my mind a very bad/flaw assumtion.... Lots of iterations for no need..
Hello Professor,
Can you please tell why and how variable angle effect on beam alignment??
Sorry, I don't understand what you're asking.
To create a constructive interference at the receiver, the processor must do some kind of perfect delay adjustment for the incoming wave. How does they do it in practice ?
Samples are generally stored in a queue for each antenna, and then it is just a matter of adding samples from different antennas with the desired time offset.
@@iain_explains Thank you professor, does that mean that the task of find the proper amplifying coefficient and proper delaying actually something called "finding the beamforming vector" right ?
Yes, that's right.
May i know why do many references consider lambda/2 antenna spacing ? i mean what not lambda or lambda /4 or any other value ?
Thanks
That's a great question. It's on my "to do" list, but I think I'll move it up the priority order.
Dear Prof, Thank you for the very helpful video. gI am studying beamforming recently. Could you tell me the difference between beam space and antenna space regarding beamforming? When performing beamforming matrix calculation, there is a difference in beam space and antenna space, but I do not know what it means.
The term "antenna space" is not one I'm familiar with. It doesn't really make sense, in terms of signal processing. "Beam space" refers to the spatial dimension (effectively the range of possible directions of departure/arrival of a beam, when using beam forming).
@@iain_explains Thank you for your reply. professor. According to the document I referenced, assuming that the channel vector is h, h is called the antenna space, and the just the 2-D DFT of this h is defined as the beam space. What does this mean?
I think you need to find some other documents, and not rely on the one you're looking at now. It doesn't make sense to call the channel vector the "antenna space", and it is not possible to take a 2-D DFT of a vector (which is only 1-D, by definition).
hello. professor. I am working as a network engineer, and I wanted to study beamforming, so I referred to the development document. Since it is an internal content, is there a difference in terminology? The document seems to calculate 2D DFT in the form of M x N channel coefficients received from a uniform square array with M x N. Does that calculation have any other meaning? From what I understand, I thought that if I took a 2D DFT, I could see how much channel gain the channel has at a specific angle, and I understood this as a beam space. Am I wrong? It's difficult because there are so many esoteric contents. Thank you for your help.@@iain_explains
Taking a 2D DFT of a M x N matrix makes sense. In your earlier comment, you talked about taking a 2D DFT of a vector - which doesn't not make sense.
Could you place the antennas 0.18m away from each other and still get the same result as the 0.06m spacing?
If you look at 5:47 he shows the pattern for 0.174 and 0.198, and basically no. But depending on what you're trying to do, maybe 0.18m spacing could do what you want in a pinch so long as you don't mind high power sidelobes pinging off in interesting directions.
I have a question pls.. how TX diversity different ? We also have multiple antennas half wave length apart wouldn't they also create a beam ? How the receiver would get multiple copies if the are already added up constructively ?
The word "diversity" can have a few different meanings. One meaning is simply to indicate that there are multiple copies of the signal. Another more specific meaning is when it indicates that the different versions of the signal are added up (for the receiver end) or precoded (for the transmitter end) according to a specific formula. This video will hopefully help: "What are Spatial Diversity and Spatial Multiplexing in MIMO?" ua-cam.com/video/MNA0xn7EeyY/v-deo.html
@Iain Explains Signals, Systems, and Digital Comms
The first meaning (different copies of the signal) how this can happen ? Since the signals will interfere with eachother and cause a beam anyway since both transmit antennas are 0.5 wave length apart
Signal don't just "form a beam" - you need to apply phase offsets in the precoder/receiver in order to "form a beam". Did you watch the video I suggested?
@Iain Explains Signals, Systems, and Digital Comms
Yes I've watched it but if we are not applying any pecoding at neither transmitting antennas and they were 0.5 wave length apart.. the applying a signal from sane source to all of the antennas without any precoding will cause the transmitted signals from all antennas to be interfering constructively (beam) with 0 degree Los to the receiver .. if that's correct how the receiver can get different copies if either it was rx diversity or sumimo
@Iain Explains Signals, Systems, and Digital Comms
I mean it will be similar to a colinear antennas array
Hello prof,
How are these cost functions of beamforming, e.g. SNR applied with optimization algorithms such as Least mean square algorithms, particle swarm optimization algorithms etc...
Just wanted to know what position should I put my router at home (home is long and narrow) for maximum beamforming benefit and learned science behind this technology. Great explanation🫡
Correct me if I'm wrong. What I understand is it doesn't really matter. With delays it can aim at any angle.
Generally speaking, yes, it doesn't matter, since the signal processing adaptively forms the beam, and can point it in any direction. More precisely, the gain and beam pattern in any given direction does depend on the arrangements of the antenna elements in your router, but most routers would be designed with symmetrical arrangements.