Developing Wireless Systems and Networks with MATLAB

Hello, everybody. My name is Houman Zarrinkoub. I'm the principal product manager of wireless communication products here at MathWorks. And it is my pleasure to be with you today in this MathWorks webinar entitled Developing Wireless Systems and Networks with MATLAB.

Let's take a look at the agenda. Following some introduction, we're going to present today's talk in about five sections. First, we're going to talk about the next generation of wireless system, 6G and its design and exploration. Then, we're going to talk about the current wireless standards-- 5G, Wi-Fi, and so on.

Then, we're going to talk about the application of AI and machine learning techniques for wireless communication and what MathWorks has to offer you on that side. Then, the what we call antenna to bits end to end simulation, the combination of RF digital and antenna and antenna array design capabilities that makes the overall design possible.

Finally, we'll talk about the notion of it's all good to model and simulate wireless systems. How about implementing it and testing it on hardware? So we're going to talk about capabilities from MATLAB in that regard. Finally, we're going to summarize what we have learned.

So this picture just gives you an overview of what MATLAB offers you in terms of wireless communication. So you see there is essentially a wheel or a kind of pie composed of three components. The upper left hand side, we call it cellular and connectivity standards, all the digital basebands and link level physical layer simulation capabilities of different standards. Could be 5G, LTE, Wi-Fi, Bluetooth, SATCOM, and so on. The functionality we provide in MATLAB for you based on existing standards.

On the upper right hand side, you see this notion of unified digital RF and antenna design. It's end to end design. When you are designing wireless system networks, you need channel models, representation of reality into computer as channel modeling. You also need antenna design and antenna essentially configuration, an antenna array. You need RF front end systems and also physical baseband signals that feed into the RF circuitry, and then essentially propagate through the electromagnetic waves over the antenna. All of that is discussed in that part of modeling simulation, capabilities of MATLAB.

All these two parts are great. But then, most wireless communication happens on a chip, on a device, on equipment. What do you have for deployment and testing? And we have multiple solutions for you. You can do algorithm testing with our software defined radio connectivity. So you have MATLAB. You connect to software defined radios directly from MATLAB.

You can do algorithm deployment and verification on FPGA and ASCIS and SOC. Essentially, we have dedicated products that makes the task of algorithms in MATLAB, lowering them and implementing them onto device easy and verification of that easy, and also production testing with RF instruments. If you have an RF instrument in your labs, we have connectivity so you can get real data. Besides the SDRs, real data from RF instruments over the air. Or send data through the air for testing.

Now, the two emerging areas that we are exploring and investing is 6G and AI for wireless. These are the areas that seems to be gaining traction. And we are providing you what we learn from our customers as tools in that area.

OK, so the goal of wireless communication, no matter what kind of wireless communication, is ubiquitous connectivity. No matter where you are in the world, you want to be connected to the internet. And you can do it by designing different types of network. Low range communication is called personal area network. Bluetooth and Zigbee and those kind of things identify that scope of communication. We have solutions for that.

Local area network, Wi-Fi, wireless LAN, we have solutions for that. We have wireless LAN Toolbox. A wide area network, cellular connectivity, we introduced the LTE as our first standard based products, then 5G. 5G has evolved to 5G advanced. We have incorporated that into our 5G systems. And we are providing you the first starting point for 6G for wide area network.

And the non terrestrial network or global area network. Satellite communication is our latest offering in 2022 and '21. And yeah, we are trying to cover, in terms of modality of wireless communication, all different standard based wireless communication systems. So every one of these things you see, we may have a product associated with that.

OK, let's go toward what we are doing with the latest and greatest, 6G exploration. So what is 6G? Right now, my phone in my pocket is a 5G phone. So it connects to the 5G air interface of the nearest base station. And people have still the 4G phones, LTE phones and so on.

The next generation of mobile wireless communication is dubbed as 6G. It will be built on the strength of 5G. So 5G is going to be reused a lot, it seems, for the description of 6G. 6G is being developed and described and specified. It's very early stages. And it envisions to provide more ubiquitous and more sustainable connectivity, meaning lower power profile connectivity of everybody and bridging the digital divide.

So it is envisioned to be a reality around 2028, 2029, 2030. We'll see what happens. And research is underway. And now, system engineering and specification phase is underway.

This is the picture I got recently from the ITU IMT 2030 document. ITU is International Telecommunication Union. It publishes the requirements for a given generation of standards early on. And then 3GPP, the Third Generation Partnership Project, usually develops essentially technology that satisfies those requirements set by ITU in IMT documents.

4G LTE was IMT-2010. For 5G, it was IMT-2020. And for 6G, the document is IMT-2030. In IMT-2020 for 5G, those areas in green were mentioned as the main use cases or the goals of 5G. Immersive communication was the notion of enhanced mobile broadband. Massive communication was the massive machine type communication. And hyper reliable and low latency communication was also mentioned as Ultra Reliable. So that was the three goals of 5G. And 5G is trying to achieve that.

There are three added in this slide in blue. Ubiquitous connectivity, 5G was doing its best for cellular connectivity. But with the addition of these LEO satellites, the low Earth orbit and satellite communication, we can go in mobile communication much beyond the range of investment in cellular infrastructure. Because a low Earth orbit can be viewed as the base station in the sky.

And even if you are in the middle of desert or in the middle of ocean or no matter where you are, if you have the right user equipment and the right satellite, you can have ubiquitous connectivity, connectivity and internet no matter where you are in the world. That's a goal of 6G.

And then, look at the two other ones, integrated AI and communication and integrated sensing and communication. Two sister technologies are merging and integrating with communication to make it better. And you will see that in our further slides. So these are the six goals of 6G as being discussed. The IMT will finalize that soon. We are in 2024. And at that time, the 3GPP will actually come and build it.

So the enabling technologies that can give you those wonderful six things are mentioned here-- AI artificial intelligence, ML machine learning. The use of higher frequency and frequencies for which there is a large bandwidth available like millimeter wave, sub terahertz, reconfigurable intelligence services. We're going to have a slide on that. The notion that in mostly urban environment or every time you have a blockage, an obstruction, if you can program your surfaces that do the reflection and refraction-- if you can program the surface, you can overcome the blockages or mitigate its effect. That's a new technology being explored for 6G.

As I mentioned, non terrestrial network, use of low Earth orbit satellites, LEO satellites to extend the reach of what Wi-Fi and cellular and all kinds of other connectivities can do. Cell free massive MIMO, the use of massive MIMO, large antenna arrays in a way that's cell free, meaning that the communication between base stations are done automatically. There are more coming up, enabling technologies. But just want to highlight these stuff.

Now, this is the timeline. We are in 2024, requirements being discussed. At some point, the ITU will have done the requirement agreement. In this particular snapshot, they mentioned 2025-2026. And then after that, the 3GPP will work. And it seems to be 2028, '29, '30 is when the requirements will be met by the specifications set forward by 3GPP.

So in response to that need to go for next generation and exploring that, we have in MATLAB as of March of 2024, we have introduced to you a new library of functions in MATLAB called 6G Exploration Library. You can get it right now, free of charge if you have 5G Toolbox, by going to mathworks.com/products/6g-exploration-library

So the name is 6 Exploration Library. It allows you to explore, model, simulate, and test candidates. They are not specified yet, candidates or your suggested 6G waveforms and technologies. You can model waveform of arbitrary bandwidth. Essentially, we take the 5G waveform, remove all the limitations on subcarrier spacing and number of subcarriers. So you can play with as much bandwidth as you like. You can do link level simulation with those larger bandwidth, smaller bandwidth waveforms plus channel modeling we provide you that support that bandwidth. And the receiver operations that we have in 5G Toolbox.

You can explore all kinds of new frequencies for RF Toolbox and RF Blockset going to areas that are being explored-- the area of FR3, the frequency range 3, which is around 10 gigahertz, is being explored a lot these days. Also, cellular frequency up to seven gigahertz and millimeter wave in the ranges of 29, 38, and higher, 60 gigahertz. And nonstop Terahertz, which seems to be also interesting, but not as much as before in higher frequencies, even beyond 100 gigahertz.

As I mentioned, configurable intelligent surfaces and AI for wireless and accelerating your simulation. Running it faster using clusters and computers and cloud. So this is what you see when you open the 6G library, as you see on the right hand side of the screen. You see examples, simulation functions and so on that lets you explore various aspects of 6G, based on a 5G starting point.

OK, for example, the 6G waveform exploration allows you to have 6G-like waveform generator function that allows you to go way beyond 275 resource blocks, each of which are 12 resource elements, right? And essentially, subcarrier spacing way beyond 960 kilohertz that is the case for 5G.

So for example, here I choose 290 resource blocks and subcarrier spacing of 3.84 megahertz. My bandwidth ends up at 13.36 gigahertz and my sampling frequency at 15 gigahertz. So you see, just by playing with those two parameters-- number of resource blocks, essentially number of subcarriers, and the subcarrier spacing-- you can explore all kinds of 6G-like signals at any bandwidth of your interest.

RIS, reconfigurable intelligent surfaces. We have solutions in the 6G library that enables you to model a reconfigurable intelligence service, and how you can change its electromagnetic properties so that it bends or it actually makes the reflected signal not be exactly as Snell's law says, the refracted angle to be different from transmitted angle. And that functionality essentially allows you to bend the signal over blockages. And using phase shifts for each element, you can constructively add interference or signals of interest and receive higher received power at your UE.

OK, so that's the future 6G exploration with MATLAB. You just go to mathworks.com/products/6gexplorationlibrary You download that for free if you have 5G. If you don't have 5G, MathWorks can give it to you for evaluation for free. But then, you have to see if you want to go ahead with that.

How about the present time wireless standards? So essentially here, showing you we have 5G as the latest mobile communication standard. And we do a lot of waveform generation on that. Satellite communication was introduced lately. It allows you to have LEO satellites and MEO and GEO satellites and all kinds of satellite communication modeled in MATLAB visually, using a wonderful visualization canvas.

We have the latest and greatest of Wi-Fi standards. 802.11be, ba, and all kinds of other standards. Besides keeping the historical a, b, g, p, ac, ax, and all that different Wi-Fi 5 and 4 and 6 and 7 and so on. And we have a new Bluetooth Toolbox, which is all standard based toolboxes that you can get and do modeling simulation based on standard based through to standard MATLAB functions.

So our customers who get the standard based products, what do you do with them? You can do signal generation or waveform generation. You can have reliable waveforms that represent a particular standard, with the bandwidth and the content and the modulation and coding and all kinds of parameterization. So you can get maximum testing done with waveforms and maximum essentially link level simulation done with these waveforms.

Now, end to end link level simulation is another use case. So you have transmitter operation specified by standard in MATLAB as MATLAB functions. We give you the channel models, the standard body that other people have developed that take this transmitted signal and distort it based on the characteristics of a channel.

Nothing in receiver is standardized. We have read lots of nice papers. And we provide you a basic receiver algorithm. So with transmitter channel model receiver, you can do link level simulation and have these kind of curves. Block error rate, throughput, packet error rate, curves as a function of difference, and other distortions of the channel. That's one of the use cases of our tools.

We also have system level simulation, meaning network level modeling, meaning multi node, multi cell, multi-UE simulations. So we have this wireless network simulation library. And you can have a 5G network composed of multiple base stations and multiple UEs. And you can run the network simulation and see how the throughput changes as the mobility of wireless UEs and devices will change.

For testing, we have lots of measurements. EVM measurement, ACLR measurement, spectral masking and so on-- this is embedded within our tools so you can qualify your signals as being good enough for receiver operations and for turning it into a product.

For those people who are interested in detecting the properties of a signal-- deciphering who is friendly, who is not friendly, which device is being corrupted and not corrupted-- by looking at signal detection capabilities, you can find out. And AI is used a lot there to see if the devices are acting properly, working properly or not, and who is in your network.

And then finally, we have tools to connect MATLAB functions and algorithms directly to either SDR, software defined radios. With one function call, you can take the signal and transmit or receive over the air from SDR. Or say, MATLAB function that creates waveforms that can go transmit over RF instruments, like a signal generator or spectrum analyzer. So we're going to talk about these use cases and customers using our standard based toolbox for.

One of the stuff we have done is instead of making you rely on learning about how to program our stuff in MATLAB-- which is great. Most people like MATLAB programming. We provide apps, easy to use apps that take the task of programming later on.

So you have an app or graphical user interface in front of you. You want a 5G waveform. You parameterize. You say, I want this kind of 5G, that kind of modulation, that kind of coding, different bandwidth. And then, you click a button.

We generate, through the app, the waveform for you. And then, we record everything you have done. So at the end of that process, we give you the MATLAB script that you just did. So you can take the MATLAB script and write a for loop and while loop around it, and make it into a large scale testing environment.

When you do that, not only do we give you the off-the-shelf waveforms that are specified in the LTE or 5G documents like NR-TMs or FRCs-- fixed reference channels-- and NR test models, but also custom downlink and uplink waveforms. Everything is up to you. Anything you can program in 5G is available to you at your fingertips.

Take a look at the Waveform Generator app. It's an app in MATLAB. You can access it through calling apps and calling this app. And the way to generate waveform is click on that button. And you see the 5G, LTE, wireless LAN, other things like satellite and Bluetooth and so on.

And you just pick one of them. In this case, I want to downlink FRC, OK? So what is the frequency range? What is the modulation coding scheme? What is the subcarrier spacing? You set all those parameters, right? And click on Generate.

We generate the IQ samples, the baseband signal, the spectrum representation as a spectrum analyzer, as well as the time frequency resource grid, symbols versus resource blocks. And we can either save it as a variable in MATLAB or generate a script for it.

As I mentioned, after you generate these waveforms, these IQ samples, complex numbers I and Q, complex numbers in MATLAB for one channel, two channel, three channel, end channels. If you have a connectivity to an RF instrument and RF signal generator, in this case Keysight Rohde and Schwartz, and leads to whatever signal generator you may have. MATLAB and 5G Toolbox connect to that using Instrument Control Toolbox. And then, you can send that from the memory of MATLAB and the device that MATLAB is installed on.

The IQ samples to the memory of the signal generator transmit over the air. On the other side of your lab, you get a signal capture using a signal analyzer. Receive over the air with all the distortions that the transmission was included. And back to MATLAB, and you decode and get EVM and block error rate, throughput, and so on.

We have done this verification. We have used hardware and software from other vendors and have received the signal. And we could actually recover the data using our MATLAB and 5G Toolbox receiver. Or vice versa. We have generated MATLAB and 5G Toolbox waveform sent over the air. And a different software and hardware has decoded what we send correctly. So we have exchanged data, and we verified that. So it's a reliable way to test your 5G Wi-Fi Bluetooth satellite with these apps and hardware connectivity.

As I mentioned, system level simulation. You have multiple UEs talking to the same base station or multiple base stations. And they send the SRS, sounding reference signals. It informs the base station on the channel representation, right? And then using the SRS, the base station will actually perform and beamform, and put beamforming on top of the resource grid so that each of these users get a wonderful multi-user MIMO signal. And the throughput is calculated for the whole cell. This is the kind of things we can do with 5G scenarios.

Wireless LAN Toolbox serves the Wi-Fi standards. We have transmitter propagation channel receiver operations, MAC frames, RTS/CTS, EDCA network level stuff, MAC stuff. Everything in one block, in one function, in one Toolbox for all your Wi-Fi modeling and simulation needs.

Bluetooth is another product as of 2022. You can do full physical layer modeling of Bluetooth coexistence with Wi-Fi and other stuff, which happens a lot in the Bluetooth LAN. Network modeling of Bluetooth, the mesh networks, the audio networks, and so on and so forth, and localization. Bluetooth is moving toward MIMO and angle of arrival, angle of departure, and all kinds of stuff. All of that in one box called Bluetooth Toolbox. And you can use that for that purposes.

One of the most loved recent toolboxes with lots of interest and lots of engagement with our customers is our Satellite Communication Toolbox. That's our communication toolbox. As you can see, it comes with a beautiful visualization of everything geocentric, of all these satellites-- LEO satellites, MEO satellite, GEO satellites-- around the orbit with the initial conditions.

And then, you run through the simulations. And you see the location of this constellation of satellites as a function of time. You do link analysis with those information or statistical waveform generation receiver design for the major important satellite communication standards like DVB-S2, DVB-S2X, DVB-RCS, like 5G, NTN, and so on and so forth. Everything in full MATLAB source code. And lets you focus on design of satellites, instead of programming basic functionality.

For example, here a customer asked us to put together a 1,000 satellite constellation with our tools. This is a snapshot of MATLAB with those tools. And the initial condition, they run. The satellites move with a certain resolution. So in this case, we have one day simulation with 60 second sample time. And we dynamically select an optimal message path between one side of the world and the other, based on the update of the satellite locations. This is one of the examples that a lot of customers want to do.

How about going toward AI for wireless connectivity, the application of AI techniques for wireless problems? We started with lots of investment in the areas of wireless communication where we think AI can play a role. Everything that's multivariate and non-linear in nature can benefit from AI modeling. Because if the number of parameters and variables becomes too high, finding out kind of a knowledge based or expertise based optimized solution becomes difficult for numerical or algebraic techniques. So AI can play a role.

Also, anything that's inherently non-linear, especially non-linear with memory, AI can get you to zoom in on the characteristics better. So spectrum sensing and signal classification, the original application. We have lots of examples in MATLAB on that. Use our tools together, either with MATLAB based AI solutions like Deep Learning Toolbox, Machine Learning Toolbox, Reinforcement Learning Toolbox, or with OSS, open source implementation of AI in PyTorch and other languages.

So we are doing spectrum sensing classification, device identification, digital pre distortion. And DPD is very important application. Optimal transceiver designs, autoencoders and decoders that the AI helps you-- depending on channel models-- reconfigure your transmitter and receiver MCS and other parameters through learning.

And localization and positioning, which is part of the sensing part of wireless communication using AI, and beam management and channel estimation and CSI, the channel state information, essentially transmission and reception with lowest amount of overhead. All the problems in wireless that AI has been applied to.

Every AI driven wireless system design has essentially four steps. You have to have lots of data. Without data and repetitive data and data at multiple levels of frequency, the outlier data, the majority of data in the average case, you need a lot of those things. And we help you, using our waveform generator apps and other things to create a lot of data.

You need AI modeling, either from OSS, from PyTorch, from TensorFlow and so on, or from MATLAB deep learning and Reinforcement Learning Toolbox. You need to train it, and then plug the AI network and test it, using data it hasn't seen before. That's the simulation and testing.

And when you are happy with your design, you want to deploy embedded devices, enterprise systems, edge cloud desktop, SOC, FPGA. And we have tools for all these stages of AI driven wireless system design. The latest examples we've given is from the 5G advanced and 3GPP on deep learning for channel estimation, for beam selection, and CSI feedback with autoencoders.

As I mentioned, you don't have to choose to do everything in MATLAB. If you have already invested in Python and PyTorch and TensorFlow and other things for AI, you can call MATLAB from Python or those environments. You can call libraries with written Python from MATLAB. Or you can package MATLAB algorithms and deployment with Python and vice versa. So all kinds of interoperability is available to you when it comes to MATLAB and Python and AI for wireless applications.

How about a class of applications that aerospace and defense customers are really digging themselves into, and they are good customers on that, the overall system design-- RF components, digital components, antenna, antenna array components. So this is essentially the ability in MATLAB on Simulink.

In Simulink, we have a continuous time differential equation solver at the base of Simulink. In MATLAB, we have essentially an index based, time based difference equation solver. So using the tool, you can actually put together an overall system that starts with bits on the left hand side, does the digital physical layer in Simulink in MATLAB using digital signal processing.

And then, you can switch to our RF functionality in RF blocks and toolbox, that enables you to be in the voltage and current domain. Essentially, do things like digital analog converters, the mixers, the power amplifiers. All these things that are RF circuitry, we have the corresponding tools.

I don't know if you know that. We have corresponding tools in Simulink and RF Toolbox and Blockset to model the RF circuitry. And then, we have tools in Antenna Toolbox and Phased Array System Toolbox. We have tools that take the antenna and antenna array and actually are in the E and H, the electromagnetic field land and model that.

So in a same model in MATLAB on Simulink, you have digital component in IQ samples and so on. You have RF components in voltages and currents. You have antenna and antenna array in E and H electromagnetic fields and channel modeling and all the inverse operation. So you can get an end to end simulation in MATLAB and Simulink.

This is an example. An end-to-end MIMO RF transceiver simulation in MATLAB and Simulink. You see on the left hand side, this TX or transmitter baseband is a MATLAB function block in Simulink that allows you to bring 5G, for example, or any other thing into a Simulink environment. Then, all these blocks here, RF transmitter, are using our RF Blockset tools in RF domain that provide you with RF front end.

Then, you go to antenna array and antenna toolbox. And you get this antenna array functionality. Then, you have a channel model that we have that operates on that domain. And then, you do all the inverse operation. And at the end, you get a nice EVM plot all in MATLAB, to see what kind of distortions you get, non-linearity, all of that stuff.

So you see signal flow, circuit envelope analysis, electromagnetic analysis, ray tracing in your channel model. All of these seemingly distinct domains have come together in MATLAB and Simulink for you to have an end-to-end reliable simulation.

One of the things that our customers have asked us and we have come through is that I want to spatial domain channel modeling. I want to put an antenna on a building. I want to specify for you the characteristics or the surfaces of the building. I want to tell you the environment and terrain.

There is a lake here. There is a grass here. There is a road here. And I want you to essentially, using RF propagation functionality, visualize the propagation of RF over real terrain area. We have done that in our communication toolbox, antenna toolbox, other tools. And you can do coverage or SINR on a 3D end-to-end map.

You can use terrain data. You can explore different propagation models. You can explore different material for your building. So all of that already available at your fingertips in MATLAB.

For example, you want to model ray tracing. Most of the future wireless communication, it's MIMO based. And you have a massive MIMO or high density MIMO situations beamforming. In MIMO, your beam width could be very small. Because if you have multiple antennas and you constructively add them together with the right phase shift, the beam gets narrower and narrower. And a beam of RF signal gets narrower. It acts like a ray, a ray of EM wave or ray of light.

And ray tracing is a way for you to see if this ray goes in this direction. It's not just a really rising in propagation. If it's array, what is my bouncing and reflection and refraction all about? And how can I model my channel using ray tracing in all directions, and come up with a channel by actually ray tracing? You have all of the functionality in MATLAB and Simulink. And you can use that with path loss.

OK, we have the last section of our talk. All the modeling and simulation and the new tools, I bet most of you didn't know we can do all of these things in MATLAB and Simulink. The RF and antenna and ray tracing and 5G and Bluetooth and so on. All that modeling and simulation in MATLAB and Simulink is great.

How about deploying it to hardware and testing it? Is there a path for us to go from high level MATLAB algorithms to low level implementation of hardware? And the answer is yes. We can go there. We can verify what you have written there, back to MATLAB. Verify it with the original design in high level implementation. And you can test it live.

I told you about that Wireless Waveform Generator app. It's a wonderful app. I want you to try it. Go get Communication Toolbox that most of you will have. And go and get the Wireless Waveform Generator app. And as of March of 2024, we have a Wireless Waveform Analyzer app. Only by Wi-Fi, but it's going to hopefully evolve in the future.

So essentially, you can take the waveform-- the 5G, the Bluetooth, the satellite communication the LTE, whatever waveform you generate in the generator tab-- then you come to transmitter tab of the same app. And the app will look at whatever is connected to your computer network.

What RF instrument, what SDR, if it's connected to your network, the app will find it. And then, the Instrument Control Toolbox or Communication Control Toolbox will connect to it directly. And you can continuously send the waveform you just generated on that RF instrument or SDR. And in the lab, have the waveform you just created be subject of testing.

It supports all waveform types-- 5G, wireless LAN LTE, Bluetooth communication. Automatic sample rate selection for USRP B/N/X and waveform sampling. And you can actually generate the MATLAB script for it. So you can actually do it offline with a script that lets you do it online or offline without interacting with it. You can actually run it overnight and come and see the results. So you connect easily to SDRs and RF instruments with Wireless Waveform Generator app.

We have this thing called wireless test bench. It's a new product. And it works only with high end USRP in an X Series. It is meant to provide you high speed data transmission and capture from MATLAB. So if you really are at the high end and you have a high end USRP X or N series with this wireless test bench, you can send a lot of high data rate and test your application. Spectral conformance, signal detection, spectral monitoring, cognitive radio, all that stuff can be done using our wireless test bench.

That's for testing. How about HDL implementation? So you have a MATLAB reference as a MATLAB 5G and so on. You add architecture to it. You add parallels and z minus 1 and delay and latency and all of that. You add fixed point to it. Right?

And then, you are ready for HDL code generation. So these collection of tools together provide you the ability to go from high level MATLAB reference all the way to Verilog HDL, code generation that can then be go to place and route and be actually implemented on an FPGA for you.

This is the path of code generation. The products involved with HDL coder and fixed point design are Simulink. And then, you can actually go and take that generated signal on the hardware with the same input in MATLAB. Bring it back to MATLAB, see if it behaved as you want it to behave. That's the HDL verification and targeting.

We have so many different applications. We have taken the RFSoC device and transmitted all kinds of stuff in there. 5G NR MIB recovery, MIB is master information block and so on. We have done all kinds of SoC implementation and so on. So all these things are available on SoC Blockset. And you can see how using Wireless HDL Toolbox and SOC Blockset, you can take an idea, an algorithm from high level MATLAB, implement the hardware, test and verify it, and be successful at.

To summarize, MATLAB and Simulink enable efficient design of end-to-end wireless systems and networks. Capabilities we have include 6G AI for wireless, modeling and simulation of current wireless standards, 5G, wireless LAN, Wi-Fi, Bluetooth, SATCOM, joint digital RF antenna design implementation, and verification of wireless design on actual hardware.

If you want to learn more, please visit our wireless communication product pages. Most of those are written by myself and my friends. So we have done a lot of due diligence to make sure that the standard specification and the MATLAB functions that we have developed for the specification are plainly described to you.

And we have a wireless communication solution page to go through different design workflow and stages from digital to RF, analog and so on, and give you the overall wireless communication picture. With that, I want to thank you and wish you a wonderful day.