Why multichannel beamforming is useful for wireless communication
From the series: Understanding Phased Array Systems and Beamforming
Brian Douglas
Wireless communication systems like 5G and WiFi usually have to serve many users simultaneously and they have to deal with multiple paths between two radios when operating in a scattering rich environment. This video covers how multichannel beamforming and spatial diversity is used to overcome those issues.
Published: 7 Dec 2022
Now that we have an understanding of what beam forming is, and at least conceptually how it’s achieved with gain and phase shifts, in this video I want to walk through how it can be used to overcome some of the problems that we face with modern communication systems like 5G and WiFi. Again, we’re going to focus on concepts in this video rather than the equations themselves, but I think it’ll still be pretty interesting so, I hope you stick around for it. I’m Brian, and welcome to a MATLAB Tech Talk.
With wireless communication systems, we are trying to transmit information from one point and receive it at another. And electromagnetic waves are a great way to do that because for one they travel at the speed of light and two we can take advantage of antenna arrays and beam forming to shape the beam in a way that benefits the communication link.
Now, wireless communication comes with some challenges due to the nature of how and where it’s used. For 1, communication systems usually have to serve many users simultaneously, and 2, they have to deal with multiple paths between two radios when operating in a scattering rich environment.
So, let’s expand on that first issue. It’s common with communication systems to have a single array that needs to communicate with as many users as possible. Perhaps the most ubiquitous example is with 5G wireless where there is a main transmitter and receiver that communicates with many different 5G radios. You know so that multiple people can be on their phone browsing the internet at the same time and in the same general location.
But this same capacity problem can be found in WiFi, where a single router is communicating with several WiFi equipped devices in a house. And it can also be found in ground-based satellite communication where a single ground station is communicating with several satellites as they pass overhead. So, this is the first thing we need to consider, how to increase capacity from a single array?
Now, one way that we can get an array to communicate with several radios at the same time is to just duty cycle between all of the users. For example, point the beam at one receiver for some amount of time, then jump to another, and then back again. In this way, we’re increasing the number of users, but at the expense of cutting the throughput from each user since it’s not a continuous link for them. We’ll see that we can actually increase capacity without affecting the throughput with multi-channel beamformers but we’ll come back to that later.
The second issue of having multiple paths between a transmitter and a receiver comes from environments where there are a lot of objects and materials that can scatter and reflect the beams. Now, this probably isn’t that much of an issue for satellite communication since there’s usually not a whole lot of scattering between ground and space, however, it is prevalent for WiFi and 5G since they tend to operate in urban environments where there are a lot of potential scatters in the form of metal sheeting in walls, large buildings, and moving vehicles.
So, the question that I want to answer is how can we take advantage of arrays and beamforming to address these two issues and be able to maintain a robust link to multiple users at the same time? Well, to answer that, let’s take a step back.
From the previous videos, we know how to affect the shape of a beam by adjusting phase and gain to each element in the array. And we know of two ways to adjust gain and phase; we could use an analog phased array antenna or we could use digital beam forming with a Multi-Input Multi-Output Array, or a MIMO array.
In block diagram form, the phased array looks like this. The RF signal is constructed that we want to send, and then that exact RF signal is sent to each antenna element in the array. Well, it’s not exactly the same since there are phase shifters for each element that are used to steer the beam and there are amplifiers for each element that will allow for gain tapering to control the side lobes. But the underlying RF signal shape is the same since it’s coming from the same RF chain. That signal is just being amplified and shifted.
For MIMO systems, each antenna is connected to its own RF chain. This means that we can construct completely different signals for each element which gives us a lot of control over the shape of the beam and like we talked about, this flexibility opens up the ability to do adaptive beam forming.
But since these digital arrays can essentially construct entirely different signals for each antenna element, we can do more than just shift phase and gain. We can combine several signals together, each weighted with their own gain and phase, to create a multi-channel beam former. This type of beamforming has the ability to construct and send entirely different signals in different spatial directions using a single antenna array. Which is important to both increase capacity and take advantage of multi-path environments.
So, let’s talk about how multi-channel beam forming is accomplished? And to start, let’s revisit single channel beam forming. We have a signal that we want to send. It’s adjusted using the weighting vector and then the signal has different gains and phases for each array element, which produces an interference pattern that forms the beam shape that we want. This is what it would look like for an 8-element array with isotropic antennas.
Now, if we have a second 8-element array, with a different weight vector for the signal, then the beam for this array could be steered in a different direction. So, if we have two arrays, we could easily send a signal in two different directions. But here’s what’s cool we can accomplish this on a single array. If we take the signals that go to the 8 elements in the first array, and add them to the 8 signals for the second array, then the interference pattern that results from this summation is the superposition of the two beam patterns.
And, now if we change the weights to the different channels, we have the ability to steer each channel separately from the other. Check it out, it’s like two search lights just scanning the sky. But it’s doing it from a single array. The key is that we have different weight vectors for each channel, and this weighting of the signals is sometimes called precoding.
Now, this is pretty cool, but it does comes at a cost. Multichannel beamforming requires a different RF chain for each array element. This is because after summing the two channels, each element has to transmit a unique signal from the others and this requires its own RF chain to construct. Hence a MIMO array is necessary.
But what I find really interesting is that both channels are being radiated out in all directions at the same on this array, it’s just that the interference pattern creates these two distinct beams. The first channel from all of the elements is constructively added in one direction and the second channel is constructively added in a different direction. So, in this way we can send the power for each channel in separate spacial directions.
Ok, so what does this mean for capacity and multi-path environments? Well, let’s start with the multi-path scenario. In scattering rich environments, there might be more than one path that the signal can take and still reach the receiver.
Now, this might not seem like a big deal since we could just figure out which path is the strongest and just point the main beam in that direction, right? I’m mean this would give us the best signal to noise ratio if we could put all of the power in the direction of the path with the least amount of loss. But, that might not always be the best solution at least in terms of providing a robust communication link. For example, in cities, the environment is dynamic. The receivers themselves might be moving, the reflectors might be moving, and there might be new obstacles that arise within the path. In this case, the communication link could drop out suddenly when the main path is blocked and then come back again once it has cleared.
So, if there is a scatter rich environment, a better approach might be to use different channels to send the same information out along different paths. For example, having an array that can send the same signal in two directions, then if both paths are available the summation of the received power could be comparable to a single channel approach, maybe a little less since one of the paths might have more loss, but if one path is gets blocked temporarily, then at least some power is still reaching the receiver through the other path and the signal is not lost completely. This is the idea behind spatial diversity. We are taking advantage of multi-channel beam forming to improve the quality and reliability of a wireless link in a multi-path environment.
Now for applications like 5G, not only do we want robust communication, but we also want to communicate with multiple users - so, we need increased capacity as well. Therefore, instead of sending the same information across multiple channels, we can also send different information on each channel. In this way if the users are separated spatially, one antenna array can communicate with multiple 5G radios at the same time by sending out different information in different directions.
Of course, with 5G, it’s usually in crowded areas like cities and so not only do we want an array with large capacity but we also want very sharp beams to target individual radios. Both of these criteria require massive MIMO arrays. In fact, 5G arrays can have dozens or even hundreds of elements.
Theoretically, an array that is this large could have hundreds of channels, each pointing in different directions and each serving a different user. However, these massive MIMO arrays introduce a new problem that we have to address.
If we go back to our block diagram of the MIMO array, we can see that in order to have multiple channels, we need to have multiple different RF chains. This isn’t a problem when the array is small, but with massive MIMO arrays we can run into a cost issue. 5G operates at mm wavelength. This means that if the elements are spaced half a wavelength apart from each other, we need to have all of those RF components packed into a very small area. This can come with some serious hardware costs which can be impractical for a lot of commercial applications.
With that in mind, we can compromise between the traditional analog phase array, where there is one RF chain and that signal is phased shifted to steer the beam, and the massive MIMO array, where there are multiple RF chains and each chain could be unique from every other one. The compromise is a hybrid approach. In a hybrid approach we can still have a huge number of antenna elements, which is important for creating sharp beams. However, we don’t need one RF chain per element, and instead could just have a few, one for each channel, which then have their own phase shifters to further steer the beams.
This approach reduces the number of channels for the array since we can only have a channel for a given RF chain, but we have reduced complexity and cost for the same sharp beam that we get with multiple elements. So, it can be a good tradeoff.
Now, in this video we’ve covered the general idea behind multi-channel beam forming. But I think it’s helpful for you to explore it a little more with some hands on examples. Specifically, I’ve linked to a MATLAB example called Improve signal to noise ratio and capacity of wireless communication using antenna arrays. In this example, it walks through a single channel communication system to provide a baseline, and then builds up the system with multi input, multi output, and multiple paths environments and shows the improvement in SNR and capacity. It’ll also help you further understand the topics that we covered in this video so I hope you check it out.
I’ve also linked to an example that walks through a Massive MIMO Hybrid beamformer. This one is particularly interesting because it also shows how you can determine the environment, or the multiple paths that exist between the transmitter and receiver, in order to create the precoding that is needed to take advantage of those paths. So, be sure to check this out as well.
Aright, this is where I’m going to leave this video. If you don’t want to miss any future Tech Talk videos don’t forget to subscribe to this channel. Also, if you want to check out my channel control system lectures you can find more control theory topics there as well. Thanks for watching and I’ll see you next time.
Select a Web Site
Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .
You can also select a web site from the following list
How to Get Best Site Performance
Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.
Americas
- América Latina (Español)
- Canada (English)
- United States (English)
Europe
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)
Asia Pacific
- Australia (English)
- India (English)
- New Zealand (English)
- 中国
- 日本Japanese (日本語)
- 한국Korean (한국어)