Design and Simulate RF Transceivers for Wireless Systems

Hello, everyone. Welcome to this webinar on Design and Simulate RF Transceivers for Wireless Systems. What you are seeing now on the screen is an AD FM comms board with an AD9361 wide range RF Transceiver chip on it.

The equivalent Simulink model for the receiver section of this RF transceiver chip would look something like this. It has AGC, a delta sigma ADC, a continuous time analog filter section, a discrete time decimation filter section, and other components, all modeled leveraging the multi-domain modeling capability of Simulink.

We can package this whole Simulink model into a single subsystem as you are seeing on the top, a single Simulink block. Let's see how we can use this block now for our analysis. So here is a Simulink model that is using AD9361 block. If I go, I click on it and go under it, I have an AGC section in it, which is modeled using state flow to model different modes of function of the AGC. And then I have a RF tunable circuit, and then an LPS circuit with analog filters, and then an ADC, and then a digital DDC circuit with a couple of half-band filters.

And again, if I go to the top of the model, I have two sources that I can connect to this block. One is continuous wave source, and then the LTE source. Let us simulate this model with a continuous wave source and see how is the input spectrum and how is the output spectrum at the output of this AD9361 block.

So what you're seeing now on the left is an input spectrum which is having two tones, one at one megahertz and two megahertz. This is the input. And the output side you are seeing in the output spectrum two tones, fundamental tones a one megahertz, two megahertz, and a DC component at zero hertz, and then an in-band noise, and then the noise floor.

We can take a closer look at the result of using this picture. The output of the AD9361 receiver consists of fundamental tones, nonlinear IP2 and IP3 components, a DC offset that got introduced due to carrier leakage, images, harmonics, in-band noise, and then the noise floor. What this clearly captures is the model clearly able to capture various effects of RF environments as the signal travels through the AD9361 RF chain.

So this is the block that I have shown you at the start. It is a result of collaborative work between MathWorks and Analog Devices. Analog Devices uses Simulink for creating behavioral models for their various RF and mixer signal product domains, like ADCs and DACs that they have, they create these Simulink behavioral models. So that one can use these models directly within their systems to integrate this model into their system and do a complete system-level simulations.

So now the question is, how can one design, model, and simulate their own RF network within Simulink. So that's what we are going to discuss in today's session. And the agenda would follow, we'll start with the budget analysis. And then we'll look at how to implement a RF model for system simulation. And then how do we integrate the implemented model together with baseband algorithms for an end-to-end simulation. And then, we'll also look at two of the application in areas like power amplifier characterization, and then the coexistence on interference analysis using this in the RF domain.

So let's begin with the first item on the agenda. So if you are someone starting with the RF design, where do you actually start? We start with the RF budget analysis by constructing the entire network and doing the budget analysis.

So if in order to make the process of the RF budget analysis easy, there is an app called as RF Budget Analyzer that comes with the RF Toolbox, which lets you create a two-port RF network by picking up the elements from the available elements in the Budget Analyzer app. And it will let you compute and visualize the power, noise, arbitrary gain at each stage of development as well as for the entire cascaded network using analytical Friis equations and as well as the steady state harmonic balance analysis on the entire network.

So unlike a spreadsheet analysis that the traditional workflow, what this app takes into account is it also accounts for the impedance mismatches between the elements in the network as well as it also take account for a nonlinearity component between the elements, so that the complete analysis will give you a more accurate results of what is my budget for the entire network.

And also, after you complete the budget analysis within the app, you can directly generate the MATLAB scripts for the entire budget, so that you can use those scripts for automation and doing some more complex simulation and analysis. Also, the Budget Analyzer app provides you with necessary options for exporting the budget analysis into RF models as well, which you can use for simulation and system integration with the other parts of the system.

So let's actually look at this RF Budget Analyzer in action. Let's say I have a superheterodyne receiver at my end. Maybe I know this is slightly blurry image. But let's say I have a receiver with components, say I have a TR switch, and then followed by a RF filter, and a couple of low-noise amplifier, and a gain amplifier, and then a D modulator that is converting my signal from the carrier frequency to the intermittent frequency of 400 megahertz, and followed by IF filters, and IF amplifiers, again followed by ADC.

So let's say if I want to analyze RF budget for a circuit like this. How do we do it in our Budget Analyzer app? Let's take a look into it. So here is a cascaded RF network built in RF Budget Analyzer app using the elements that are available in the Budget Analyzer app. And the parameters for this each element, say, what is a gain, noise figure, an IP3 point for each element, well, you can specify on the left side for each of them. And then at the top, you can see the input system parameters, what is the input frequency, or frequency of the input signal, and the power, and the bandwidth of the input signal.

So and on the results section, you would see first analysis result. As you keep adding an element into the cascaded network here, the Friis results will pop up. The Friis result includes the power output gain, noise trigger, IP3, and SNR at each stage of the cascaded network as well as for the complete cascaded network.

Not only the Friis analytical equations, as I told, we can also perform a steady state harmonic balance analysis. You can do it so by clicking on the HB-Analyze button at the top. It stimulates the entire network for a harmonic analysis and populates the results as you are seeing it in the below. So it also performs a similar things what we did with respect to Friis analysis. But again, giving a more accurate results are taking into account of nonlinear effects as well and analyzing it in a frequency domain.

Well, along with the Friis and HB-Analyze options, RF Budget Analyzer app also provides you with multiple options for visualizing the parameters at each state of the cascaded network. You can plot the transducer gain at each stage. And for the cascaded network, similarly you can plot the noise figure of the network as well. And also, use options for plotting as parameters of the entire cascaded network. You can use Smith charts, XY plots, or a Polar plot to visualize as parameters for the network.

And once you do and create the entire cascaded network, and finish your budget analysis, and make sure that whatever the budget result of the cascaded network is coming-- is meeting your requirement, then you can use this Budget Analyzer app to export this into a-- there are multiple options. One, you can export it into rfbudget object into workspace. Or you can export this as a MATLAB script as well. So let's export it into MATLAB script.

So here is the RF budget cascaded network that got exported into you MATLAB script. As you see here, you have all the elements that you modeled in the Budget Analyzer app. And under the below, you have the rfbudget object that will get created when we run this script.

Now, let me run this generated script. And once I run this script, you will see the rfbudget object that got created and came into the workspace with a variable name b on the right here. So you can just give the variable name to visualize how is the rfbudget object appear. So it says that here are the input specifications, and here are the complete elements that are captured into single rfbudget object.

So once you have the budget, if you would like to again open it in the Budget Analyzer app, you can just give a show b, and it will open up the rfbudget object into the RF Budget Analyzer app again for you to do further budget analysis and continue with computing the complete cascaded network analysis.

So we have looked at the first section of the RF design part, which is creating the entire cascaded network and doing the budget analysis by leveraging the capabilities of the RF Budget Analyzer app. Now, let's move into the next section. So we have fixed the network. Now how do we do it, the simulation of it, in a time domain to be precise? Before we, again, discuss in RF simulation, let's start with what are the requirements for simulation.

The first requirement is RF system level simulation should be fast. So if we look at different options available for us for modeling and simulating the RF network, one at the lower end of the spectrum, we have true pass-band simulation, which would simulate the entire frequency network and captures the RF behavior of the network along the entire frequency range. We call it a [AUDIO OUT] transient analysis of the RF networks. We captured the entire behavior. But the problem with this, in order to do so, the time-step required is very small. Which means the time required for entire simulation to finish would be very long.

And on the other end of the spectrum, where most RF engineers this is an equivalent baseband kind of simulation, wherein we try to capture the RFB only at the fundamental or the carrier frequency. So while the time-step for this equivalent baseband depends only upon the bandwidth and gives high simulation speed. But the problem is, it cannot capture the RF behavior at the other frequency components.

A trade-off between these true pass-band simulation and equivalent baseband simulation is what we call it as a circuit envelope simulation. Whereas in circuit envelope simulation, we simulate-- we capture the RF behavior at the fundamental carrier as well as with the-- at the harmonics of this carrier, and also say DC, 2f1, 3f1 like that.

So circuit envelope is the one the solution for us in order to take better advantage in terms of the modeling fidelity as well as in terms of getting a simulation speed. And so, to achieve the fast simulation, we'll be using circuit envelope simulation in our simulation process.

And the other requirement for the RF side, the models that we use to model this entire RF network, it should be accurate and should be able to capture all the nonlinear effects, noise generation, any frequency dependency should be clearly represented using the blocks, and also modeling of noise and modeling of impedance mismatches between the network should be handled properly. So the two requirements, the simulation should be fast. One solution that we thought for it is circuit envelope simulation, and the model should be accurate.

The answer for both of these things is what is called as RF Blockset, which is an add-on to Simulink. And this RF Blockset product uses a circuit envelope simulation technique, what we have discussed in the previous slide, and will enable us to simulate linear, nonlinear, and noise behavior introduced by these RF networks. And again, this RF Blockset models will be running within the Simulink itself.

So as it is still in the Simulink platform itself, it is very easy to integrate the RF networks with the remaining part of your system. Say you have a baseband part, which is having some signal processing algorithms. So you can create the RF network model or the Simulink model for it and integrate together with the baseband processing algorithms to simulate the entire network in a single platform. I'll be showing that more in my upcoming slides as well. So here, we use RF Blockset in order to do this circuit envelope simulation.

So now, how do we connect with between the budget analysis that we've done and creating this system level model for circuit envelope simulation? How do we do that? The process is very simple. As we saw in the first stage, we create RF budget analysis on entire cascaded network. We export it into an rfbudget object. RFB here is showing the rfbudget object with all the details of this cascaded network.

So once you have the rfbudget object, you can create rfsystem object from it, which will be the Simulink model for you. So you just execute rfsystem of the rfbudget, and it will create the equivalent Simulink rfsystem.

So to give you in more detail about what I was explaining, let's take an example. Let's take direct down conversion as an example. On the left, you can see the RF Budget Analyzer screen, where I am doing a direct down conversion modeling of the cascaded network. Say I have a parameter filter, amplifier, and just a direct demodulator, and then the amplifier.

So from this, I finished the budget analysis, and I export into rfbudget object. And from rfbudget object, I quickly create the rfsystem object. So this system object will give you the corresponding Simulink model, which you can use for the Simulink simulation.

Again, when you are working within the MATLAB, you can directly use the system object variable, object variable name itself to pass your inputs and gets the outputs, which will at the back of processing, it will call the RF Blockset, do the circuit envelope simulation, and it will give you the outputs. Say, so if you are having a network like this, you pass the input to the rfs object, and you collect the I and Q outputs at the output of this.

And let's say, the rfsystem object will also facilitate you for further heralding the Simulink model as well. Say if you created an rfsystem object. You can open the Simulink model by using open underscore system of that rfsystem object. Then it will open up in the Simulink model like this. Now you can either directly run the circuit envelope simulation within the Simulink and obtain the result.

And also, you can further edit this Simulink model to introduce, say, whatever the nonlinearities that you think you would like to model further. You can see, in this picture we have a modulator with all the block parameters, so you can configure phase noise and then the other nonlinear parameters into this block and can run the simulation to observe it. So that means rfsystem will help you to directly do the simulation within the MATLAB or export it and do it in the Simulink as well.

And I would also like you to point out you the one of the new features that got introduced in the latest 2022 year release. Wherein now rfsystem object can help you to automatically generate MIMO systems from a single RF budget chain. So what I mean is, let's say you have created an RFB, rfbudget object from a single cascaded chain-- single two ported network cascaded chain.

But if you want to use it in a-- if you want to use the same RF chain for creating a multi-MIMO system, wherein it has multiple RF chains, you can use rfsystem off by specifying how many chains you want. And it will automatically create the model with the number of chains that you specify, so that you can directly use it in your model for simulating the multi-chain network.

And let's discuss about the blocks itself. I know this slide might look a little overwhelming. But let me tell you that the RF Blockset offers a rich set of circuit envelope blocks, which helps you let you model, design, and simulate any kind of RF transmitter and receiver network. It has multiple linear elements like say, attenuators, couplers, circuit, phase shifters, and then the nonlinear elements like amplifier and the mixer.

Again, starting with the R2020a, you would be able to model amplifier using AM/AM or the AM/PM table data that you get from your vendors and mixers using the IM intermodulation table data that you generally get for a mixer from the vendors. And there are also blocks available for modeling thermal and phase noise and blocks for modeling the variable networks, like say VGAs, variable attenuators, and variable phase shifters. So these can be used in modeling RF transceiver sections like ADC, beamforming, Butler matrix, or wherever there is adaptive kind of nature in the RF network.

And then, there are system blocks like modulators and demodulators, which can help you to model the effects like carrier leakage, image reduction, channel selection within during the upconversion or the downconversion process of the signal. Again, I have showed you exhaustive list of blocks here. But again, if you have a block that isn't part of this available blocks, you can order your own blocks using the Simscape language. Please reach out to us. We can help you in that as well.

And then, there are other than the blocks that are also a kind of testbenches that are provided using the RF Blockset. So how these testbenches help is, they help to generate the stimuli and measure the response from a DTE-- sorry, a DUT, a device under test. So how do we use it? Let's say I want to measure an IP3 point of a cascaded network that I built. So what you can do is, you can pick up the IP3 testbench, and you can pick up your network, connect them, and then you can run this model to observe the IP3 point of this network.

And again, this IP3 testbench has a block parameters wherein you can tune the stimuli that you are generating. So you can change the input frequency, change the input power levels, and basically tune the stimuli and observe your corresponding IP3 point for the entire network over the range of input values.

And again, so I have shown what's available and then the testbenches available. So and also to help you get started quickly with the modeling of some of the RF effects, like say phase noise modeling, thermal noise modeling, harmonic components like IP2, IP3, or the intermodulation distortion and the image reduction ratio at mixers, there are reference examples available, which will help you to get started quickly and do the analysis.

So we have covered the parts of how do we start with the building a design using budget analysis, and then how do you create a Simulink model or a RF system model for circuit envelope simulation using the rfsystem objects. Now, we have the RF Front-end. Now, how do we integrate and simulate it together with the baseband and antenna so as to do an end-to-end system-level simulation?

So if you basically look at a wireless system, at a very high level, you can divide it into three components. One, the baseband where you can design your algorithms, generate waveforms, and perform measurements. And then the RF front end, which handles the digital to analog conversion and the up and down conversion to the carrier frequencies. And then the antennas and antenna arrays that would convert the electrical signal into electromagnetic radiation for transmission intersections. And then, there is channel, though that is not inherently part of the system, it is necessary when you are doing a system-level simulation in order to understand how are the channel affects between the transmitter and receiver.

So another point that I'm trying to address is, we have the RF front end that we model. We have the blocks, and we are also doing the circuit envelope simulation to analyze the effects of the RF front end. Now, how do we bring in this RF front end together with the baseband and antennas so that we can have a complete very good level of fidelity model for simulation? So let's start looking at it.

So let's look at an end-to-end MIMO transceiver simulation including the antennas and the beamforming. So where do we start? Again, we can start with the budget analysis for a single chain in the receiver for the RF front end. So what I'm doing here, I'm modeling a receiver front end chain with an antenna element, amplifier, and a directly down-converting to the baseband frequency amplifier, followed by I'm using a phase shifter here, which I'll be using for doing the phase shift beamforming at the receiver.

So I'll do the budget analysis. I'll check if all the budget within my requirement. And then, I will export it in the rfbudget object. And then, from the rfbudget object, I will use rfsystem to create the Simulink model for the receiver RF front end.

So if you see here, this is the complete receiver model. And here, we can see that it is a MIMO receiver, that means it's a eight receive chain. So while creating the rfsystem, we will give rfsystem of eight receive chain, so that it can directly create an eight RF chains from the single rfbudget object.

So we now have a receiver RF front end. And I would also like you to draw your attention to this specific block which is showing the multiple antenna arrays. Maybe I talk a little on that as well. So this antenna array block that you're seeing in the model, it's a new block that got introduced in, again, that 2022 year release. So what this will help is, you can design an antenna or an antenna array using an Antenna Toolbox. And you can import that antenna object directly into an RF network using this antenna array object.

So what do you do by doing this is, you are bringing the effects of say, what is the radiation pattern, or what is the polarization of the antenna. You bring these effects together directly into simulating with the RF network by creating the antenna object and importing it into this using the antenna array element. It also specifies whether you are using a transmitter configuration or a receive configuration. And you can also specify the frequency of operation and the direction of arrival or direction of departure.

Not only that, so when you are matching your antenna structure with the RF chain, you might be having an impedance mismatch in between the RF chain and then the antenna element array. You can also model that by specifying the S parameters corresponding to that impedance mismatch in the modeling section of this block.

We have looked at the modeling of RF front end chain for the receiver. Now, let's look at the modeling of RF front end chain for the transmitter part. It's again, the process is similar. Here you are using a phase shifter to, again, do the beamforming on the transmit side, or direct modulator for upconversion amplifier followed by a transmitting antenna.

And then again, you use rfsystem object with of eight transmit chains to create an RF front end chain for transmitter. And again, you are using a TX array here to model the complete transmitter chain. Now, what we have is that we have the RF front end receiver chain and RF front end transmitter chain. So the next step would be to combining everything, the transmit front end RF chain and then the receive front end chain into a single Simulink block.

So if you see here, you have everything combined here together with the free space channel in between. And once you have the complete Simulink model starting from the signal generation, to the front end, and then to the receiver, and then to the baseband side of it, you can simulate this model to understand how is your received signal is getting affected through all the impairments.

So if you see, it has multiple components in it. It has a TX baseband. It is currently generating an OADM signal. And then the front end that is modeling the entire phase shifting, modulation, upconversion, and then the nonlinearities through amplifiers as well. And then, you have antenna array that includes the effects of radiation pattern, polarizations, direction of transmission, and reception. Similarly on the receive side. And in between transmit to receive, here I am using a very generic of free space pass-loss.

But if you are someone would like to model and use a specific channel model, you can use it. And not only channel model, you can also bring in the environment effects, like a AE4, how do the transmission affect that is there. So what we are doing, we are building a system-level model that has completely a very high level of fidelity.

So [AUDIO OUT] you can change maybe your direction of transmission or arrival and see how is my receive signal is getting affected. Or you can introduce more power amplifier back off in the transmit intersection, or introduce, say, more noise floor and see how is my detections are affecting at the receiver.

And also, we're having the complete model while you are designing the receiver algorithms at the baseband level. Also, you are taking every effect into consideration. So that when you translate this into your implementation, you would have already considered all the effect. So it would be ideally meet the requirements that you would face in a real-time scenario.

And one more point that I want to highlight from this slide is that, as you see, the signal starting from the baseband signal generation, it is flowing through multiple domains. You start with the signal generation, a signal flow domain, and then it is doing RF circuit envelope stimulation at the RF front end, and then bringing in the EM simulation, electromagnetic wave simulation effects at the transmit antenna array. And again, going-- traveling back the path to, again, to the signal flow domain.

That means it's the signal is going through multiple domains. But again, everything is now in a single platform Simulink. So that's how Simulink enables for doing your multi-domain modeling across the system for system level simulation. And I think this is one of the key takeaway slide I would like the participants to, say, snap it as a key takeaway from this session.

And this actually brings me to establishing MATLAB and Simulink as a unified design platform for entire wireless system, Be it be baseband, RF, antenna, or modeling the channel, MATLAB and Simulink can act as your single unified platform for end-to-end system. And here, I am highlighting the to say, the corresponding products which would help in modeling the particular parts of the wireless system.

So again, what I have shown when I'm showing the combined TX RX system is a more generic transceiver. But again, if you are in some specific application areas, we do also provide a reference architecture model for that. Let's say, if you are doing a hybrid beamforming that involves the analog beamforming and also the digital pre-coding matrix and all, we have the examples to get started with.

Similarly with the, if you are introducing-- plan to introduce RF front end into the 5G NR waveforms. Similar with case with the radio broadcasting or a ZigBee network that you might be working. And if you are someone from the aerodefense space, you might be interested in modeling RF front end effects to your radar system simulations. Yeah. And we have a reference architecture available for you to help you get started on that quickly.

Yeah. So we have discussed the three parts of our agenda today. One is budget analysis, and then using the simulation-- doing the simulation using our RF system, and then integrating everything together with baseband and antenna, doing an unturned system. Now, in the latter part, we'll discuss on two of the application areas or highly interested areas in this domain. One is power amplifier modeling, and the other one is interference analysis. Let's start with the power amplifier modeling part.

We have to start with a bit of theory. Here is-- we will try to understand how digital pre distortion helps to linearize the power amplifier. So power amplifier generally has this-- exhibits nonlinear characteristics, as indicated by this compression orange colored line. But what we need is a more linear characteristics.

And this is where we bring in the digital pre distortion, with the characteristics that are inverse of the power amplifier. So that when a signal passed through the combination of this digital pre distortion amplifier, they would cancel each other, and we would be expecting to get a linear PA performance.

Yeah. And the theory, this sounds very simple and straightforward. But when we go into practice, it becomes more complex. Because one thing is, power amplifier characteristics are not only nonlinear but also involves memory effects. So what does this mean by involving memory effects is, the output at any point in the power amplifier also depends upon the previous inputs as well, not only just the present input, but also the previous input.

And further, the behavior of the power amplifier depends upon the input signal modulation, what is the bandwidth, what is the power, what is the peak-to-average power ratio of the input signal. So whatever the digital pre distort of that we are designing, it has to be adaptive, and also has to operate in a closed loop fashion with the power amplifier.

And again, to make the things more complex, the digital pre distorter is something that we keep it in the digital domain. Whereas in a power amplifier, it is an RF domain. So it involves upconversion and down conversion and also need to take great care of the timing alignments between the things. Otherwise, we end up in an unstable situation. So again, this is where we might need a platform like Simulink for doing an end-to-end simulation that handles both the baseband part as well as the RF part.

So first, let's start with understanding how to, if you have a power amplifier, how do you bring that power amplifier into Simulink for circuit envelop simulation? It's very simple. You get the first power amplifier raw data. And then, you try to fit the data with the memory polynomial. So that whatever the output that you get from the power amplifier when you are giving the input, you should get the same output from the memory polynomial when you are giving the same input.

And you verify that in both time and frequency if you are getting the same output from that memory polynomial or not. And once you see the match where you collect the memory polynomial coefficients, which you will be using in the circuit envelope simulation. So maybe I will provide more detail in terms of those here.

So what I was telling is, you collect the PA data. And then MATLAB provides you a white box fitting procedure that will fit power amplifier to your memory polynomial. Again, this is completely open box. You can read and edit the code as well. By fitting, you collect the polynomial coefficients. And then we supply these coefficients into a power amplifier blocks that we use it in the circuit envelope simulation. So you specify the coefficient that you get in the coefficient matrix of this power amplifier block.

So let's look at a very recent example that got introduced, again, in the latest release that talks on power amplifier and DPD modeling and for dynamic EVM measurements of 5G waveforms. So what we-- this example actually details and explains in detail all the process involved in characterizing the power amplifier and also designing the DPD and doing a combined simulation with the DPD + PA.

So firstly, we use a R and S instrument to get the power amplifier characterization data. And here is an application note from R and S site, indicating that this is a, let's say, collaborative work, and indicating how a simulation environment like MATLAB and Simulink work hand-in-hand with the physical devices to collect the measurement data.

So you collect the measurement data from the power amplifier. And then you use the white box in a white box fitting process that I showed in the previous slide to do the fitting of the power amplifier to a polynomial, and collect the polynomial coefficients, and then bring the polynomial coefficients into the memory and power amplifier block that we are using here.

And surrounded by this power amplifier block, we build the entire baseband network as well. So here, you can see the DPD coefficient estimate block that will estimate the inverse coefficients that need to be fed to the DPD. So this DPD coefficient estimator block is available as a Simulink block. And this can estimate the coefficients required for DPD by using least square of the recursively squared of estimation. And then this block will do the DPD correction on the incoming signal and then pass it through the power amplifier.

So once you have the block complete system-level model like this, you can simulate and-- how do you say, analyze the performance effects. So without DPD and with DPD. Similarly, you can simulate this model to get a different constellation plots, how is my cancellation affecting with dividends without DPD, and then the corresponding EVM results as well. So if you are someone working on the power amplifier site, I would highly recommend to try out the latest example in 2022a to get complete idea of the analysis process.

So that brings me to the last part of this presentation, which is a multi-carrier simulation for coexistence and interference analysis. If you're-- I believe you might have heard about the latest industry news, where the newly installed 5G base stations causing interference to the radar altimeter causing some of the airlines to stop their flights.

The problem is simple here. You have a radar altimeter band. And with a radar altimeter-wide filter. Whereas you have the-- when you transmit the 5G signal, you have a fundamental frequency. And then the 5G spurious emission that is coming into the radar wideband filter and affecting the detections of the radar altimeter.

So MathWorks has published a very detailed blog post on it with the title "C-Band 5G Telecom Delays and Airline Frustration," which describes in detail about the entire scenario and how you can model the same scenario within the MATLAB and Simulink, and analyze the effects of 5G and radar interference. So again, with the system, so growing with the technology and coexistence within the nearby bands, it's more than ever required to do a interference analysis and coexistence analysis before we deploy it.

Quickly, I would like you to point out to one of the examples. So here, we have a 5G. Here we have a Simulink model that has a 5G NR waveform that you are generating using a wireless waveform generator. And then, you are also generating an LTE waveform. So what we are doing in this example is, we are passing the both the 5G waveform on the LTE waveform. And we would like to see how is this LTE interference to the 5G network is impacting our 5G decoding.

So I have a, again, RF network which is, again, to operate on a receiver front end with all the nonlinearities model. And this RF receiver front end, we can specify what are the carrier frequencies. So the 5G NR signal is coming at around 2.19 gigahertz an LTE carrier frequency is coming at a 2.12 gigahertz. The receiver network takes the RF front end, the model using Toolbox and Blockset, will take care of bringing the effect of LTE onto the 5G.

And on the baseband side, you try to decode the signal and analyze different things, like what is the AC PR, adjacent channel power ratio, or adjacent channel leakage ratio. How is my condition affecting, and how is my EVM getting affected? So it's all like, once you have the system-level model for the entire thing, it's very flexible to analyze the interference analysis by leveraging the multi-carrier simulation within the RF front end networks.

So before I go for a poll question, I would like to draw your attention to a couple of stories, one from the Qualcomm and other from the nano-SIM. Both of them use an RF Toolbox and RF Blockset for their RF transceiver development. And the results are that as they code, the development time get reduced by almost 50%, and then it highly accelerates the verification process, and also makes the verification very easy and comfortable.

So here, I would like to-- the next RF design, and get rapidly started with RF Budget Analyzer app to do your cascaded RF budget analysis, and then use rfsystem objects to directly create the Simulink models for RF design and similar simulation. And also, leverage the multi-domain modeling capabilities of MATLAB and Simulink to integrate your RF front end together with the baseband signal processing algorithms to design more adaptive RF networks.

And then, you can also model MIMO systems, including the effects of antennas and other power amplifier nonlinearities into the system, so that you can simulate the entire whole network as a single code with a very high fidelity.

So again, if you would like to exploit full potential of this products, we have a more on-hands training available, which is RF system design using networks tools. This is a two-day training course. Please reach out to us, and we would be happy to guide you on the steps for registering to this course. And it details how do you design an RF networks using the tools, and how to simulate, and how do you create the custom models as well.

And to give out the summary, you might have seen this particular pie chart during this webinar series a couple of times. So this is a three-part webinar series. And in the first part of the webinar, we have touched up on wireless connectivity standards. And here are the list of products.

And in second session and in the current session that we are discussing, in the second session, we focused on machine learning partpart in 5G, and in the current session, we discussed on how do you build RF front end for your wireless system, and how do you integrate that with the baseband and antennas, and bring out the MIMO systems, making MATLAB and Simulink a unified design platform.

So again, we have covered a bit of the prototyping and testing But you could expect more on prototyping and testing in the latter part of the webinar series that we are going to connect the next half of the session. You could soon expect the communication from our end on this next part of the webinar series where we cover most of the prototyping and the testing part.