A Detailed Introduction to Beamforming
HTML-код
- Опубликовано: 3 авг 2024
- An introduction to Radio Beamforming, including the basic mathematical expressions that allow to predict the how antenna arrays behave.
Just plain trigonometry and basic linear algebra is used.
How arrays are able to focus radio signals on the target and filter-out interference. 
Наука
![Jordan Adetunji - KEHLANI REMIX (feat. Kehlani) [Official Video]](http://i.ytimg.com/vi/EeJ8n5PxFGE/mqdefault.jpg)
I think, I just wasted my money in university for these courses. You have explained this concept to the point and very clear.
Thanks for your comment.
I am very blessed by this video..I really appreciate this and may God richly bless you as you move forward to make more videos for the benefits of upcoming learners. ❤
Thank You. Very well laid out topics with animations. It actually feels worth of more than 2-3 1 hour lectures of university. Cuts the time, yet delivers the information with much more insight. :-)
Thank you!
Damn this is one of the most detail explanation of beamforming. I have taken this class before but barely able to pass
Thank you, keep up the good work. This was very well explained, simple to follow and very informative. Content like this is what makes RUclips.
Thank you!
Thank you for the excellent presentation: clear, concise, and to the point. Thanks!
Thank you very much!
We were searching for beamforming, it's good, keep it up
Thanks!
Those examples and simulations are good, especially with the clear explanation.
Thanks!
Very detailed. Complex subjects explained in simple words. Impressed. Expect more videos from you. Thanks.
Thank you!
One can only describe this as s top notch brilliant presentation. The presenter respected the audience, where, rather than offering a hand written scribbled page which he himself read, while wavering a pen or a pencil on a hastily drawn sketch, he showed that he was a highly responsible person, who respected the subject and gave it the decor it warranted. Thank you for this video.
It always interested me to know whether within the volume of the formed lobes or in any other radiation pattern, do the E/M curled loops of half a wavelength in size, stack above and near each other as they propagate out from the transmitter.? A one megacycle will have large E/M loops but a 5 Giga hertz signal will have small E/M loops. This suggests that stacking of E/M loops must occur as in waveguides and cavity resonators working at the higher modes. where they need to stack " half wavelengths E/M loops above and beside each other. No one seems to describe this activity as the wave propagates out of the antenna and most drawings show the whole lobe or ever growing in size E/M loops which cannot be sustained due to the limitations set up by the permittivity and permeability of the medium to restrict the rate at which the E/M field grwo and decay at the frequency they are launched. One would suggest that in phased array antenna the E/M loops on the periphery of the lobe are phased out while the other in the centre of the lobe remain stacked up with higher power intensity. It would be interesting to discuss and find an algorithm describing the actual stacking of individual E/M loops within the resulting lobe or in other antennas including omnidirectional ones. Stacking is never mentioned in antennas only in waveguides and cavity resonators. but I am confident that "stacking" occurs in antenna propagation.
I learned the detail concepts and mathematical expressions from your video. Thank you for your class
I am glad to know that it is useful. Thanks for your comment.
Super dude... I was using array antenna without knowing it properly for calibration of different antennas... Now i understand it's importance..I like very much in this video is demonstration of radiation pattern in spectrum analyzer...
Thanks!
thank you for the presentation, it was very clear and straight to the point
You are welcome.
Great explanation! Thank you.
this helped me "gain" a great intro to beamforming!
Super good video very well explained! Thanks!
Thank you very much!
Wonderful presentation, thank you very much for this.
Glad that you liked. Thank you!
This is so clear. Thanks so much
You are welcome.
Congratulations by the high-quality,,
Thank you!
Lovely explanation! thanks
Thank you!
excelente explicación , hice un repaso de antenas cuando la estudie en la University of Carabobo -Venezuela
Muchas gracias, Arnulfo.
Really useful.
I like your fundamental approach. Greatly explained. Please upload video related to MIMO.
Thank you!
Your video solves most of my confusions from university's lectures
Glad to help. Thank you.
@@5glearning772 I have some questions.
1. Is Beamforming only applicable for receiving signal as in your video. How about transmitting signal?
2. Your video takes the examples of up-link. How about down-link case? Do they implement "antenna array" on mobile devices and apply beamforming there?
Thank you so much !!!
1. Yes, beam-forming is also used to focus the radiated energy for transmission.
2. In the downlink, the mobile device can be equipped with an antenna array to enhance reception. But in practice only a small number of antennas can be embedded when the operation takes place at cm wavelength.
Current 5G devices include 4 reception antennas, so you may consider that a modest reception beam-forming exists at the device. In practice, these antennas are used for MIMO reception (this is another topic).
In the uplink, transmission beam-forming could be implemented at the device.
But current devices transmit with only one antenna at the same time due to cost-saving considerations (transmission components are relatively expensive and drain significant energy from the small battery). This situation may evolve.
If operation takes place at mm waves( e.g 29 GHz), it is feasible to embed a substantial number of antenna elements in the device. Therefore, beam-forming at this side may be used and certainly helps to compensate the high propagation losses at these frequencies.
I hope that this answers your questions.
Excellent !!!, thank you very much
You are welcome!
This was very really helpful for understanding my academic syllabus
I would like to see some other vedios Please do
Glad to know that it is helpful!
Thanks a million
You are welcome!
Hi , good explanation thanks
Thank you!
thank you so much
You are most welcome.
Thank you.
You are welcome.
Hi, sir, i have problem to calculate gain function of antenna when we use beamforming technique, for that please help me sir and thank you.
Thank you for the amazing introduction for beam-forming, One little confuse!!, at the moment 18:11 of the video regarding beam steering, the X-axis is (-45 to 45) degree!! here I understand that the reception is only happen within 90 degree! or should the x-axis changed to (-90 to 90) degree?
In this figure the numbers 0, 15, 30, and 45 represent gains (in dB) instead of angles on degrees. In this sense, you should consider these values just as relative amplitudes of the array gain with reference to an arbitrary basis. You are right stating that this simulation covers the angle spanning between -90° and +90° from boresight (i.e. half a circle).
Reading your comments, I realized that I should not have included the numbers: -15, -30 and -45. They don't have physical meaning and I will correct the figure in this respect, when having a little time to edit the video. Sorry for the misunderstanding.
Thanks for explaining the phase array antenna. What is the equation needed for array gain during 11:47 to 12:14? I used the Array factor equation but could not generate the plots during the period 11:47 to 12:14.
Thanks for your comment.
Following are the equations to compute the array gain as a linear expression (you may take logartihms to express it in dB).
Let N be the number of array elements, d the distance between adjacent elements, and lambda the wavelength.
The array gain depends on theta.
If theta =0, Gain = (1/2)*N^2.
If theta is not 0, call
Num (N) = 1 - COS(N*2*PI*(d/lambda)*SIN(theta))
Then, the array gain is given by
Gain = (1/2)*Num(N)/NUM(1)
Hope this helps.
Excellent. At 3:58, for equation y(t), the modulation component x(t-T) is replaced by x(t). Please explain. Thanks
It is an approximation applicable to narrow-band communication systems.
In this video, the underlaying assumption is that the bandwidth of the modulating signal x(t) is much smaller than the carrier frequency w0. This is typically the case in 5G systems where, for example, bandwith is on the order of 100 MHz for a carrier frquency of 3.5 GHz. This is what we mean by "narrow-band". Intuitively, the modulating signal changes at a much lower rate than the carrier.
In a typical antenna array, we select T (or tau) as a fraction of the carrier period (typically not more than one half). Therefore, we can assume that the modulating signal keeps its value quite constant during this short delay.
In example at time 2:18 the wavelength of 1 GHz is given in correct value. Please note
Hello, I understood all thank you.
My question is : Can I configurate 4 beams by 8 antennas??
Hello. Yes, you can. There are many alternative ways.
A simple approach would be to split the antennas set in pairs, and use each pair to configure one beam. It would work, but not very efficiently because the lobes would be relatively wide, so you would get low pointing accuracy.
On the other extreme: (1) you could include an adder at the front-end of each antenna element, and (2) you could compute 4 sets of 8 weights, one per beam. Then, you would get relatively narrow beams in each direction at the expense of significant hardware cost (or higher DSP load, if beam-forming implemented in SW).
There are intermediate solutions, as well.
Thank you for excellent video... Shall we use this MIMO concept in Radar Technology?
Thanks for your question, and sorry for the late response.
Yes, the MIMO concept is used advantageously in radar. An excellent book dedicated to this topic is:
MIMO Radar - Theory and Application
Jamie Bergin - Joseph R. Guerci
Artech House
ISBN-13: 978-1-63081-342-0
All these technologies in telecom are broadly speaking borrowed from defence technology so telecoms is like few decades late but mainly because digital beamforming etc is now affordable and minituarized in chipsets for mmwave technology.
Hello, thank you for the great and clear presentation. I have a question regarding array factor. According to the presentation, the gain of array factor can be calculated analytically. It has closed form solution. How can we predict the total array gain (gain from array factor + gain from single element )? I see a formula saying total array gain = number of element * gain from single element but I don't think it is correct. A gain of infinite small dipole antenna is around 1.5 in linear. Placing this dipole with separation lambda/2 does not give the total gain 1.5 * 2 = 3. My question is how to predict additional gain from array factor? Thank you.
Thanks for your question.
To obtain the array gain, the radiation pattern of a single element must be multiplied by the array factor, but note that both functions depend on the observation angle (theta in 2D, theta and phi in 3D).
On the other hand, adding elements to the array leads to several consequences, one of them being a general increment in gain that is proportional to the number of array elements.
Using the array factor is an easy to use approximation, but observe that we are neglecting the influence on each element of the other ones. This is usually called "mutual coupling". If we need accurate results, a better option is to compute the array gain using the "active element" approach, in which we consider the radiation from a single element and evaluate the resulting (passive) radiation from the others.
Following your question, I ran some simulations using this last method for two infinitesimal dipoles, and - yes - I observed a difference.
Despite that, the simplistic "array factor method" remains a good way to understand the physical mechanism behind phased arrays.
I hope this may help...
5G Learning Thank you fro the tests and explanation. I actually compared the full array result with other two methods you mentioned here. I realized the approximation holds only if the directivity of a single element is not high. This means the directivity is not significantly different between single element and active element pattern. Max{G_array} = N * Max{G_single} can hold. It is also elaborated in Array Handbook 2rd edition page 74.
@@s0914025 It is good to know the validity limits of this approximation. Thanks for the observation, and for the link to this excellent book!
Thanks for the video. Can you please make video on MU-MIMO
Thanks for your comment. I am starting to build a video on MIMO, including multi-user, but it will take some time...
Hello, may I use part of this video in a slide of my training material for staff training purpose only?
Yes you may. Thank you.
hi, very good video. Can you please tell me if I want to design my own Phased Array architecture, Can I use Matlab to simulate it?
Thank you very much.
Yes, Matlab is an excellent tool for that purpose. The "Phased Array System Toolbox" includes many functions and examples that would greatly facilitate your task.
If this toolbox is unavailable, you might always develop your own models from scratch. You would get a deeper understanding, but it would also take much longer.
Besides, if you would like to build actual hardware, the "Antenna Toolbox" can simulate radiation diagrams of simple or crossed-dipoles, and many types of patch antennas.
Hi sir, superb video. I have a doudt, what parameters need to consider, in order to decide a weight formula, what is weight exactly
Hi Suresh, thanks for your comment.
A weight is a factor that multiplies the signal received from one of the antenna array element.
Suppose, for example, that your antenna array includes 5 elements. Then, you need to consider 5 weights, one per antenna element.
To process the array signal in reception mode, you add the weighted signals from each antenna element. Technically speaking, you perform a "weighted average" of the signals from the array elements.
A similar approach applies to transmission mode.
The weights may be complex numbers, i.e. including modulus and argument.
The higher the modulus, the more significant will be the signal from the corresponding antenna element.
Besides, the argument is used to introduce a phase-difference (i.e. a delay) between the signals from each array element.
There are many ways to select the weights.
Some of the methods allow to steer the array sensitivity in reception mode, or the radiated power in transmission mode. These methods are collectivelly called "beam-forming".
Other methods select the weights to increase the array sensitivity towards the target and set a null in the direction of an interference source (or a jammer).
Explaining the detailed procedures to compute the weights takes some time, so unfortunately I cannot delve into this subject in this limited space.
I hope that this helps.
Hi, First of all. Thanks for uploading this lecture. This was very informative. At 11.05, the figure notation follows spherical coordinate system. Then why did you said to adopt cartesian coordinate system? I am new to beamforming. I am a little bit confused.
Thanks for your comment. You are right, (r, phi, theta) are spherical coordinates. Sorry for the confusion.
Thank you for the excellent explanation. Where can I get the matlab code for array factor of linear antenna array for random amplitude and random phase.
Thanks for your comment.
You might consider the following book: "Practical Guide to the MIMO Radio Channel with Matlab Examples".
Authors: T. Brown, E. De Carvalho, P. Kyristsi
ISBN: 978-0-470-99449-8
Another alternative could be chapter 29 of the book "Problem-based Learning in Communication Systems Using Matlab and Simulink"
ISBN: 978-1-119-06034-5
Authors: K. Choi, H. Liu
I hope that it may help.
Thank you very much
@@adityajvs You are welcome.
hi what are the simulation software that u used to shown all the antenna array performance in the video?
No special simulation tool was used for this video, just Excel.
@@5glearning772 where u guy obtain the array pattern result @12.52
@@boonyang88, it is an application of the array factor equation described in the previous slides. An array of 8 elements is considered. For this animation, the modulus of this factor is presented using logarithmic scale.
Excuse me at 16:00, why is there two delays considered?
good video, although I have a doubt, in the minute 21:50 to 21:60 mathematically why the signal is blocked or attenuated by 30 °
Thanks for your comment. The signal arriving at 30° is attenuated because it is not wanted, it is assumed the interference. Therefore, the weights are computed to attenuate this signal in order to better receive the other one, arriving at an azimuth of 0°. I tried to show through this example that an array can be used to "filter-out" interference. This is used in the receiver of 5G base stations. It is also used in radars.
Awesome video! Where can I find more info? What other resources can I read? 😄
Thank you!
An excellent book is "Antenna Arrays - A Computational Approach" by Randy L. Haupt. ISBN 978-0-470-40775-2.
@@5glearning772 Many thanks!!
Hi! Are You sure about the lambda in 2:00? It's 30cm for 1GHz, not 10cm.
hello! nice video. I would like to know if there is a way to exploit the array of antennas to see if a normal Bluetooth signal (from a smartphone for example) comes from the front or from behind the receiver without using Aoa, Aod and CTE signals.
I know that Bluetooth 5.1 has these possibilities but I can’t use them because I need compatibility even with previous bluetooth version.
I don’t need a great precision just to know whether front or back.
Thank you very much !
Thanks for your comments.
I am not familiar with the details of the Bluetooth standard.
In general, current smartphone antennas are quite omnidirectional. Therefore, it is difficult to assess where the signal comes from without an array.
Arrays embedded in the smartphone will be common when millimeter bands become available (it is already happening in some countries).
In the meantime, you may consider an external antenna with a metallic reflector. it would help to identify if the signal comes from the front or from the back.
I hope that these comments may help.
@@5glearning772 Thank you for your kind reply. I was thinking of using two omnidirectional antennas separated by a metallic reflector and see from which antenna the connection takes place. It could work ? Do i need two bluetooth modules ? Thank you anyway !!
@@emanuelebagalini3913 Consider that you place the reflector at a quarter of the wavelegth from the antenna.
This reflector will increase the antenna gain in the direction opposite to the reflector. Signals arriving from this side will be perceived at a higher level than if they would come from the other hemisphere.
But your reflector having finite size, the waves from the other hemisphere will somewhat reach the antenna through the physical mechanism of diffraction on the reflector borders.
Hence, you should use a reflector as large as your setup allows.
Another alternative would be to place a directional array at one side of the reflector instead of a basic antenna (e.g. a dipole), therefore increasing the overall gain in the intended direction.
Besides, if your target is static or moving slowly, you could just use a singe receiver, switching cyclically its input between the two antennas.
I hope these comments help.
@@5glearning772 "Sorry , I don't want to bother you too much, but I have another question. I looked for an answer all over the internet, but nothing came up. In a scenario with two devices, for instance a powerful bt module (200mt/650ft range) and a smartphone (10mt/30ft Bt range), I would need to detect if the smartphone is in the range of the module. In order to detect the smartphone I have to send an inquiry signal and get a feedback containing its ID. Now, my concern is about the antennas. Since I have the transmit power to send a long distance signal, doesn't it necessarily mean that the smartphone should be able to reach me back from 200mt am i wrong ? Because of his weaker transmit power. So, I think I need also receiving antennas but don't know if they have a proper "range". In conclusion I have the "volume of voice" high to make me hear, but not the "ears" good enough to hear the answer is that the case ? Or did I create a problem out of nowhere? Is a comunication possible ?
Thank you for all your replies, you are really helping me !
@@emanuelebagalini3913 Let us call "1" your smartphone and "2" your powerful BT module.
Communication 2 -> 1 is feasible at long distance due to the high transmitting power of 2.
But communication 1 -> 2 may not be possible because the signal from 1 might arrive at 2 at a very low level, as you accurately observed.
Therefore, in order to facilitate 1 -> 2 a possible solution is to increase the gain of the "2" receiving antenna to compensate for the lower transmitting power from 1.
For this purpose you could use a directional antenna, but if you don't know a priori where 1 is, you are would be unable to steer your directional antenna to the desired direction.
A second alternative would be to use an antenna array as discussed in the video, instead of the directional antenna mentioned above.
Using a sufficient number of antenna elements, the gain may be as high as wished.
Following this approach:
(i) you can obtain the high-gain required to compensate the low power of the smartphone;
(ii) you can electronically steer the antenna, scanning over a 180° or 360° azimuth range (like a radar).
Hello,
I really liked the video. Can you share the slides please ?
Thanks
Thank you. For the moment I prefer just to publish contents in video format.
Hi! Can you please let me know the software that you used for the beamforming simulations you had shown in the video? I mean, is there an easy-to-use and user-friendly software available? If yes, kindly let me know. Plus, did you use an actual acoustic array for coming up with these simulations?
Hi, the simulations that you see are deduced from the basic equations of antenna arrays. I have just used Excel for computation and graphics, although Matlab would be suitable as well.
Besides, these are only theoretical calculations. No attempt has been made to build accoustic, radio or optical experiments because, unfortunately, I have currently no time to work in this direction.
Hi,
Can anyone tell me please how variable angle affects on beam alignment??
Hi,
Alignment accuracy is higher if the target is near to boresight direction (zero degrees azimuth for a 2D antenna).
If the target is located far from this direction, the accuracy decreases because the width of the main lobe inevitably increases.
In practice, 5G antennas are typically designed to cover an angle of 60° at each side of boresight, with an accuracy on the order of 4°.
I hope that this answer your question.
Are there some new videos comming up ? I have an idea , please make a video for Angle of arrival calculations and extraction of useful information by fft of received signal and spatial rotation of the signal to even further reduce the data points , channel estimation for UL/DL , with this data
Thanks for the suggestion.
Currently, I have no time to produce new videos.
@@5glearning772 may i ask what are you working on now a days ? just curious
@@mubeenamjad3619 I design and test 5G networks.
Make a video on use of aerial basestations in 5G
Hi, may I translate this video to Indonesian language (Bahasa) ?
Yes, you may. Thank you.
@@raulbruzzone2525 Thanks :D
can i get matlab code for phase shifting antenna
Not yet, sorry. Publishing code requires additional time and resources that I lack at this moment.
I download it before the exams
Good luck in your exams!
Phased Array Antenna pls
where can I get the codes?
As the simulations from this video are relatively simple, I just used Excel.
To translate the equations into code, I would suggest Matlab or one of its clones (Octave, Scilab, etc.).
There is also a specific Matlab library dedicated to this subject: Phased Array System Toolbox. It is optional.
With this library you can reproduce all the simulations described in this video and much more. It is very easy to use.
Yay! We all live in phased array and can beamform ourselves on demand! What could possibly go wrong? Ever seen the movie "The Dark Knight"? You know the part at the end where Batman uses everyone's cell phone in Gotham to view all of the city in real-time at once? Then Lucius Fox resigns because it's "unethical" and "dangerous". Yeah, you will have no privacy whatsoever in a 5G smart city. Every move you make has and is being tracked. Yay for "security"! The scene in the movie is called "High Frequency Generator".
Can you elaborate more on this? I mean, I have not watched that movie.
But you know if you can just write a bit more in detail....
Your accent is too funny that I can nearly impossible to focus on the content.
This guy doesn’t seem to understand the topic and is just reading from a textbook.
Thank you for the excellent presentation: clear, concise, and to the point.
Thank you very much!
thank you for the presentation, it was very clear and straight to the point
Thank you.