Interference and Coexistence of Bluetooth and Wireless Systems

Hello. Welcome to this webinar on interference and coexistence of WLAN and Bluetooth systems. My name is Rameez Ahmed, and I'm part of the Application Engineering Group at MathWorks and focus on the wireless standards-based products, including 5G, LTE, WLAN, Bluetooth, and satellite communications. I have over 12 years of experience with wireless communications and have been with the MathWorks for about eight years, spending the first few years in communications and digital signal processing toolboxes development.

In our webinar today, we will talk about how WLAN and Bluetooth systems can coexist in the same physical environment and operate in the same band. Here's a brief agenda for our presentation today. We will start with a brief introduction of what the terms "interference" and "coexistence" mean. We will then move on to introducing the WLAN and Bluetooth Toolboxes that are essential to model the two systems. We will then drive our attention towards modeling interference at the PHY level, including topics like adaptive frequency hopping, and then we will drive towards a more system-level simulation, where we have multiple nodes, and talk more about collaborative and noncollaborative coexistence.

So let's start with our introduction. Over the last few years, the concept of ubiquitous connectivity has become very prevalent. So ubiquitous connectivity describes a state where the devices are able to create, share, and process data and never lose connectivity. It enables uninterrupted connectivity between any devices in the inner environment.

Now, this ubiquitous connection is achieved through radios operating in different protocol stacks, so for your personal area networks, you've got your Bluetooth and Zigbee networks. For your wide area network-- sorry, for your local area network, you have your Wi-Fi. For wider area network, you've got your cellular standards, which is the 3GPP, including 5G and LTE. And then for your global area networks, you have satellite communication standards, such as the DVB-S.

Now, this combination of all of these networks is what truly achieves ubiquitous connectivity, and given that these networks have to exist and operate in the same environment, there is going to be interference. So let's talk about interference. What is interference? Interference occurs when two or more radios operate in the same frequency bands.

Let us take, for example, the 2.4 gigahertz ISM radio band. We have many devices that operate in this particular frequency band, including Bluetooth devices, Wi-Fi, your old cordless phones that we used to have until a few years ago. That have become pretty extinct right now. You have baby monitors right now that operate in these bands. Many devices operate in these bands.

And due to the ubiquitous deployment of these wireless networks and devices on these 2.4 gigahertz bands, multiple homogeneous and heterogeneous networks, including Wi-Fi, Bluetooth, Zigbee, are likely to coexist in the same physical environment. Now take, for instance, this Bluetooth channel. This shows, basically, the 2.4 gigahertz Bluetooth LE channel. It's basically 40 different channels, each with a channel bandwidth of 2 megahertz.

You're also seeing another Bluetooth, Classic Bluetooth, the basic rate and the enhanced data rate channels. These are 79 channels, each with a bandwidth of 1 megahertz. Now, these can coexist in the same environment as a WLAN system that follows the 802.11a, b, c, ax, be protocols, and these can coexist in the same environment. So what happens when a Bluetooth signal that can frequency hop hops on to a channel that has a WLAN present in it? You end up with an interference scenario.

Now, the concept of coexistence is when two of these systems have to stay in the same environment and exist together. Now, this can be achieved through two mechanisms, one as collaborative coexistence. Collaborative coexistence is where the nodes acknowledge the presence of other nodes in the network and can share information between them in order to take any mitigation measures.

So here's an example of a smart phone that is connected to a Wi-Fi router, headphones, and a fitness watch. Now, in this scenario, you can see that all of these three-- the links exist in the same physical environment, and the phone can-- and can perform certain mitigation measures in order for all of these links to coexist. The second type of coexistence is noncollaborative coexistence. This is where you have links that are not colocated on the same device and don't really have any communication between them in order to coexist in the same environment.

So here is an example of that. There is a laptop that is connected to a Bluetooth speaker. You could have a cell phone that's connected to a fitness watch, and you could have a television that's connected to a Wi-Fi router through a Wi-Fi link. Now, in this case, what can you do about coexistence, and how can you take mitigation measures? So that's what we're going to talk next. What are the mitigation measures that can be taken by these nodes?

So let's talk about the collaborative coexistence mechanisms first. The first method that we are going to discuss is alternating wireless medium access. So in the AWMA mechanism, a WLAN radio or a Bluetooth-- and a Bluetooth radio are colocated in the same physical unit, which enables a wired connection between these two radios. The collaborative coexistence mechanism uses this wired connection to coordinate access to the wireless medium between the WLAN and the Bluetooth radios.

The second method that we are going to discuss is packet traffic arbitration. In the PTA mechanism, the WLAN station and the Bluetooth devices are again colocated, and the PTA control entity provides per-packet authorization for all transmissions. This mechanism can deny permission for transmissions if it has a chance to-- if there is a high probability for a collision. Now, the PTA mechanism dynamically coordinates sharing of the wireless medium between the traffic loads of WLAN and the Bluetooth systems.

The third method that can be used is the deterministic interference suppression. Now, in this mechanism, a null is inserted in the WLAN receiver at the frequency of the Bluetooth signal. Because Bluetooth devices hop to a new frequency for each packet transmission, the WLAN receiver must know the hopping pattern and the timing of the Bluetooth device. The hopping pattern and timing is obtained by using a Bluetooth receiver as part of the WLAN receiver. Now, deterministic interference suppression is a colocated collaborative coexistance mechanism.

So let's talk about the noncollaborative mitigation schemes and starting with adaptive frequency hopping now. The simplest one is we all know that Bluetooth uses frequency hopping, so the adaptive frequency hopping enables Bluetooth to adapt its environment by identifying fixed sources of WLAN interference and excluding them from the list of these available channels.

The second method that we can use is adaptive interference suppression, so this mechanism is exclusively used-- it's related to signal processing in the WLAN physical layer. It works by requiring a Bluetooth receiver colocated with the WLAN receiver, and without any prior knowledge of the timing or frequency used by the Bluetooth network, the WLAN receiver will use adaptive filters to estimate and cancel the interference signal.

The last method we are going to discuss is adaptive packet scheduling. Now, this is basically-- Bluetooth transmissions involve various packet sizes, various packet properties, including packet length, degree of error protection used. Now, by selecting the best package type according to the channel conditions for an upcoming frequency hop, networks can achieve better throughputs. Additionally, packet transmissions can be scheduled efficiently so that the Bluetooth devices can transmit during the hops that are outside of the frequencies and refrain from transmitting while they are in-band.

So that's a whole bunch of-- that's a brief introduction to the interference and coexistence and how we can mitigate and a whole bunch of different methodologies that have been used by these two networks in order to mitigate the coexistence. So in the next section, what we're going to talk about is, how can you model some of these scenarios in MATLAB? And before we can go and start talking about that, we need to talk about the WLAN and Bluetooth Toolboxes, so let's go ahead and do that.

So let's start with the WLAN Toolbox-- what is the WLAN Toolbox? The WLAN Toolbox provides PHY and MAC layer modeling capabilities for the 802.11 standards. They can range anywhere from the 802.11a, b, g, n, ac, ax, all the way up to be. All of these standards are covered. And it provides transmitter propagation channel and receiver capability. So one of the most basic capabilities that is required for some of these modeling is to generate standards-compliant waveforms, and we will talk in a little bit about how you can generate WLAN and Bluetooth standards-compliant waveform from these toolboxes.

In addition to the PHY layer, we've also covered the MAC and MAC frames and the MAC implementation in the WLAN Toolbox. What this enables is measurements such as latency and throughput using the WLAN Toolbox. On a similar fashion, we've got the Bluetooth Toolbox. The Bluetooth Toolbox has full PHY layer modeling, including the Bluetooth Classic and Bluetooth LE. We've also got all the different modulation and error-control schemes available, and the standard specify transceiver tests have been implemented as well.

One of the main capabilities of the Bluetooth Toolbox is to be able to model coexistence and some of its mitigation measures that we talked about in the introduction section. The Bluetooth Toolbox also includes network-level modeling just like the WLAN Toolbox, including creating piconets with arbitrary number of nodes, creating and running a Bluetooth mesh network. So you can explore some of those routing protocols, and you can also assess the performance of an LE Audio network from the Bluetooth Toolbox.

Now, finally, we can also do localization and positioning with the Bluetooth Toolboxes using things like angle of arrivals, angle of departure, and tracking objects using sense effusions and also estimating the range between two Bluetooth devices. Now, we've talked about the toolboxes, but what are the primary use cases of these toolboxes? How do most people use these toolboxes? And the first and foremost use case is signal generation.

We want to make signal generation, waveform generation as simple and as easy as possible. Once you've generated those waveforms, you want to be able to take those waveforms, put them into end-to-end simulation scenarios. So this is where channel models become important, and we'll talk a little bit about channel models in our next couple of slides.

The third bigger use case is when you have a-- when you have a Bluetooth or a WLAN system and you have a device under test, so let's say, a power amplifier or a digital filter, and you want to see what kind of impact the digital filter or the power amplifier has on the modulation characteristics of your waveform, you want to go and do measurements on them, including things like EVM, ACLR, spectral masks. And this is where the measurements really come into the picture.

And the fourth use case is signal detection, so this is where you take-- you bring in a waveform into MATLAB, and you want to able to detect and demodulate these waveforms. And this is also enabled with our hardware radio connectivity, so if you have a software-defined radio or a signal generator, for example, a test and measurement equipment, like or Keysight equipments, you can bring in waveforms from them into MATLAB, decode, and demodulate them. And finally, system-level simulation-- so this is where we go from beyond just two nodes. We go to multiple nodes, and you want to measure performance in terms of interference and also a network-level throughput.

So let's talk about the individual capabilities really quickly. What about waveform generation? We've got a Wireless Waveform Generator app that will enable you to generate these waveforms quickly and as easily as possible. Here's a screenshot of that app, including-- that's showing WLAN waveforms, and you can see that it's a fully GUI interface. You can just do a couple of clicks to generate this waveform. It also has visualization built into it.

In addition to that, it also has a Transmitter tab where you can select from a whole host of TNM equipments and also software-defined radios in order to able transmit these waveforms into the air directly. So you can go from nothing to a fully functioning transmitted waveform in the air in just a couple of clicks, and these are fully standards-compliant waveforms. Currently, we support 5G, LTE, WLAN, Bluetooth, and some SATCOM waveforms in the Wireless Waveform Generator. In addition to that, we also have some radar and just some custom OFDM waveforms in the app.

So let's talk about the channel models. What channel models are available in here? So in the WLAN Toolbox, we've got the WLAN-specific channel models, including the TGn, ac, ax, and ay channel models. So these are stochastic channel models that have been defined by the standard and implemented in the WLAN Toolbox.

But in addition to that, we also have raytracing channel models in case you want some more deterministic channel models. So here's an example in the Communication Toolbox that has raytracing, and you can import STL files for certain indoor environments, place transmitters, and receivers in exactly the places where you want them to be and then do a raytracing to get a channel model out of them.

In addition to that, the Communication Toolbox also lets you model multipath fading channels, additive white Gaussian noise channel models. In addition to the channel model, you can also have RF impairments included in your simulation, so if you wanted phase and frequency offset, phase noise, some amplifiers, or some IQ imbalance, those are possible.

Now, you may be wondering-- you've got the waveform. You've got the channel models. What about receivers? By all means we've got receivers both in WLAN and the Bluetooth Toolboxes, and for example, if you were to take 802.11ax and you wanted certain simulations to be performed, we've got multiple examples that will go from single-user packet error rate to a multiuser MIMO OFDM, trigger-based uplink packet error rate, beamforming compressions. So different scenarios have already been implemented as shipping examples that you can take and run. Now, one of the beauties of these toolboxes is that they are all open, editable MATLAB code, which means that if you wanted to go and dig into the deep leaf-level code, those are all available, not blackbox, easily accessible code that you can just open and run.

The same is true for the Bluetooth. You have access to link-level simulations, end-to-end packet error rate be bit error rate performance. You can have receivers that are practical receivers that compensate for RF impairments, including carrier and frequency offsets, timing offsets, DC offsets. And in addition to that, you can model many variants of path losses, including free space, lognormal, shadowing, log distance, two-ray ground reflection, and the NIST PAP. These are different path loss models that have been proposed from the Bluetooth sig.

So now that we have an understanding of the WLAN and the Bluetooth Toolboxes, let's go and take a look at what is-- how we can model that interference and coexistence scenarios, and let's start with the PHY layer because we want to understand what happens at the PHY layer and then go and build on top of that and go to the-- go to a system-level simulation.

So let's start at the PHY layer, and we're going to walk through an example right now because what better than seeing an actual example? So we're going to go do a PHY-level modeling of interference, and in this we're going to be-- just to set up the stage for you, we're going to be evaluating the effect of an-- what effect an interfering wireless node, otherwise called IWN, has on an affected wireless node.

So we're going to have one link with an interferer in them, so we're going to access a Bluetooth link with an WLAN interference. And ideally, we want to see what sort of impact the interference can be in terms of space, where you have a link, and then the interfering nodes are there. What sort of impact does the interfering node-- when they're close to each other, when they're farther from each other, what sort of impact does it have?

And what impact do these interference have in terms of time? So for example, here you could have a full collision, in which case an packet from the affected wireless node is fully interfered by an interfering node. You could have a partial collision, where only a part of the packet has an interference. Or you have just no collision, where they are perfectly separated by time.

Now, since we talk about space and time, you also have the same thing in frequency. Now, here's where you want to study the impact of them being on the same frequency channel. Or they could be separated by the frequency channels, but then what sort of impact does the spectral leakage-- what impact does it have on the other systems? So we want to be able to study all of this.

Here's what we are going to be doing in this demo. We're going to be generating a waveform. So we're going to have an AWN transmitter. We're going to add some RF impairments into that signal, model it through a path loss, and then we're going to have two interferers and each of them with their own path loss so that we can mimic certain distances. We're going to add some additive white Gaussian noise and pass it through a receiver model to compute the packet error rate. At this point, we can also classify channels so that we can feed the frequency hop sequence back to the transmitter so that they can perform some adaptive frequency hopping. So let's go and take a look at that example in MATLAB.

So here is the MATLAB main desktop, and we are going to steer ourselves to the examples that we can open and run it. The easiest way to do that is by going to the doc command. Doc brings up the documentation page for MATLAB. And we are going to look at Wireless Communications, and specifically, the examples that we are going to look at are part of the Bluetooth Toolbox.

Now, under this, you can see that they have been categorized into different categories, and the category that we are going to look at today is coexistence modeling. And we're going to go into the Examples tab so that we can directly run a example. The specific example that we're going to first run is the PHY layer simulation of Bluetooth and WLAN coexistence, so let's go ahead and run this script.

So here's the Bluetooth example that we want to run today. Let's take a look at what what's in the example, and then we will run it once with adaptive frequency hopping and then another time without the adaptive frequency hopping. So here's the same explanation that I had for you before about doing the interference that are separated in space, time, and frequency, and so here are some of the parameters for the simulation itself, so what kind of system we want to simulate. Is it a Bluetooth LE, BR, enhanced data rate?

And then we also get to choose between having a turn on and off the adaptive frequency hopping. Here are the two interfering nodes that have been defined. The first one is a WLAN system. Now, you can also either generate these WLAN waveforms from a selection of these waveforms if you wanted to just click on them. In this case, we are going to do the generation with a baseband file.

You can have this waveform coming from a file, or if you have the WLAN Toolboxes, you can generate them on the fly. In addition to that, we are also going to generate a second interfering node that's going to have a Bluetooth system that's going to be the interferer. So let's go ahead and define some specifics for the environment. We're going to use an outdoor environment. We are going to define a 10 dB SNR for this, and we're going to be simulating it for 500 packets.

I'm going to skip some of the intricate details of these simulations in the interest of time, but remember when I said earlier that you have access to all of this code? Here's what I meant. If you were to double-click some of this and just click Open, you will get the leaf-level code, so if you wanted to make modifications to this, if you wanted to go and explore your own algorithms or just do a plug-and-play, these are all things that you can do from here.

So let's, again, quickly walk through the example. We go ahead and determine the appropriate path losses, and then once we've set up the scenario, we go ahead and run the simulation for different number of packets. We can talk a little bit about the Bluetooth waveform generation right here. We've got the BLE waveform that's being generated, and then we also have the Classic Bluetooth waveform that's being generated right here.

We go ahead and introduce some timing offsets, frequency offsets, and DC offsets. These are the RF impairments that we wanted to introduce. And once we've done that, we go ahead and add the interference and add the noise to the system, and then everything from here is the receiver portion of it where we go ahead and build a practical receiver and demodulate it and get the packets back from there.

We're finally going to plot some results so that we can take a look at some of these metrics. So let's first begin by running the simulation without any adaptive frequency hopping. So we've turned off the frequency hopping, and let's go ahead and run the simulation. So there's the spectrum and spectrogram plot, and so this is the results without the-- without any adaptive frequency hopping. And so let's go ahead and take a look at what the packet error rate is.

We can see that the bit error rate obtained is about 3.8 e to the power minus 4. So let's go ahead and repeat this experiment only now turning on the adaptive frequency hopping so that it can hop off and avoid the interfering channels. So let's go ahead and turn it on and run. Now we can see that the adaptive frequency hopping happening right here and with a certain pattern, and the Bluetooth system is now able to hop off and avoid the other interfering signals.

And at the end of the simulation, we will also see what the new packet, the bit error rate, looks like, so once this is completed, we will take a look at the bit error rate to see if it has improved over the 3.8 times minus 4 that we were looking at earlier. So there it is. With the adaptive frequency hopping, we can see that the bit error rate is indeed lower, at 8.5e minus 5. So we can evaluate that adaptive frequency had an effect on the bit error rate of the system.

So let's go back to the slides now. So we just saw what happens at the PHY layer when we have adaptive frequency hopping and how this can be modeled in MATLAB. The next section that we're going to focus on is system-level modeling, so this is where we take the simulation that we just ran and extend it towards a multinode network. And then we want to also evaluate both collaborative and noncollaborative coexistence. How can you perform these operations here?

Now, in order to do that, I want to introduce the wireless communications network modeling library, so this is a free add-on that you can download and that enables you to model network-level simulations within MATLAB. And network, ideally, talking about them, these are a group of nodes that are communicating with each other, and these could be anything from a star network, where you have one transmitter or multiple receivers. You could talk about an infrastructure grid. You could have mesh networks, or you could have a hybrid mesh network, where you have multiple bands involved, and we've taken the task of implementing all of this in MATLAB using our network modeling library.

Now, how do we perform network simulations? We define nodes, and these nodes can be either a Bluetooth or a WLAN node, and within each node we have the different layers, including our application layer. It could be a MAC or RLC layers. You could also have PHY, the actual physical layer transmitters and receivers and then the buffers.

And then during each iteration of the simulation, we figure out if there are any packets that need to be distributed and then move the simulation forward to the next possible invoking time. And one of the additional aspects of the simulation is also that you can perform certain actions that can be scheduled during the simulation. Say if you wanted to add an extra node into the simulation or you wanted to move the position of one of those nodes during the simulation, these can be performed as actions that are scheduled actions during the simulation.

So like I was saying, coexistence analysis can be performed at different levels of fidelity. You could have it as just one network reacting to the interference from an external signal, so here's an WLAN signal that is interfering with a Bluetooth piconet. You could talk about networks interfering and reacting to each other, so this is the noncollaborative, and then we also can talk about collaborative coexistence.

So let's begin with the first one, which is modeling a Bluetooth network with just a WLAN interference. Now, this is a shipping example in MATLAB, and I'll point it out to you when we get there. But you can basically simulate Bluetooth and LE and Classic networks in the presence of WLAN interferences. You could perform adaptive frequency hopping for those Bluetooth Classic and LE. If you wanted to plug in your own channel classification algorithm, you can do that as well, and then you can also compute the packet error rate statistics with and without interference.

Now here's an example. Here's a picture of that classification example that we just talked about, so each of these individual channels have been listed here. So you've got your 39 channels for that Bluetooth Low Energy, and you can see that wherever there is a WLAN interference in there, those channels have been classified as a bad channel. And so the same thing repeats here, and then all the other channels have been classified as good channels.

And you can also see a measure of the success rate that you're seeing, so you can see that whenever there was a WLAN interference, the success rates have been dropped significantly, whereas whenever you did not have the WLAN interference, you can have-- you usually have good performance in here. So this is an example of having a Bluetooth piconet with the WLAN interference and some channel classification algorithms to go with it.

Now let's talk about noncollaborative coexistence. I'm not going to run this-- in the interest of time, I'm not going to be running this example, but this is an existing example inside the Bluetooth Toolbox where you can configure a network with WLAN, Bluetooth, and then the Classic Bluetooth nodes, and you can measure the traffic performance for varying levels of Bluetooth and WLAN traffic.

This is an example of a noncollaborative coexistence where the only thing you can do is probably adaptive frequency hopping. I'm more interested in the collaborative coexistence case, and so we will be running that example. But before we even go there, one of the newer features of the network-level simulation is that when you perform these a network-level simulation scenarios, you can still capture IQ samples from these network-level simulations, and you can use these captured IQ samples to analyze your receiver designs. If you want, you can use it for feed hardware for testing, or you can also use it towards training and validating some of your AI or machine learning algorithms. So that's a noncollaborative coexistence.

So let's talk in detail about collaborative coexistence, and we'll walk through this example. So collaborative coexistence, like I said before-- the mitigation measures can be a couple of different methodologies, and in this particular case, we are going to be using packet traffic arbitration. You can also experiment with your own PTA algorithms, and you can also have antenna isolation for those separate antennas for these systems.

So let's go ahead and run this example in MATLAB. So the second example that we are going to run is this collaborative coexistence of Bluetooth LE BE, the basic rate and data rate, and WLAN using a PHY packet traffic arbitration. So let's go ahead and take a look at what's inside this example, but before that, we're going to open up this actual example.

Right now I'll walk you through it a little bit, and then we will run it. So like I said, the example is based on having a packet traffic arbitrator, so you've got your Bluetooth device and a WLAN device and a PTA that arbitrates this traffic. Now what does the PTA do? Here are a set of rules that the PTA will follow, starting with if a node wants to transmit multiple transmit packets simultaneously. The PTA will select the packets to be transmitted, and if a node has two incoming packets at the same time, the PTA decides whether the packets are processed by the Bluetooth or the WLAN device. Essentially, it's just an arbitrator, and here's a bunch of rules that it will follow.

So let's start by looking at some of this-- how this simulation has been set up. First check is the wireless network support package. Like I said, this is a package that you can download. You download this package by going to the Home tab in MATLAB and looking under Add-Ons. And doing Get Add-Ons. Search for network simulator and-- sorry, the network library, and you will be able to download the communication library for that.

Once you've done that, we go ahead and initialize that network and then define the different nodes that are going to be part of the simulation. So we start with a WLAN station with a device configuration of-- that have been-- that has been defined here, so we're going to call it a station. And it's going to use the 2.4 gigahertz bandwidth and channel number one.

Now, the next devices that we're going to add is the central BR device. We're going to define the central AE device, and then we are also going to define the peripheral devices. And once we are done with that, we are going to go ahead and associate all the appropriate nodes so that they know who they are connected to, and after that, we're going to add the application-layer traffic.

So in this case, we're going to use just on-off traffic for the WLAN source and the coexistence nodes. So we repeat this process for the BR traffic and the LE traffic, and once we have done that, we go ahead and attach the packet traffic arbitrator. Again, you have the option of either choosing to have the arbitration or not. You can run the simulation multiple times with and without the packet traffic arbitrator.

We go ahead and add the nodes to the simulation and then just go ahead and run the simulation right here. So let me go and run this so that you can see some of this live in action. So here's a picture that shows the packet communication over time and frequency. If I were to zoom into one of these specific nodes, you can see a lot more information about the packet transmission. So you can see things like the contention period, when packets were received, when packets were transmitted, and you can see this for the different nodes, including the coexistence node that has all the BR, LE, and WLAN node together and then each of those receivers on the other-- all the other nodes that are they are connected to.

You can also get the same information in the Frequency view. You can see that the channel number one for the WLAN node is right here. That has all the WLAN traffic. And in addition to that, you've got all of the other Bluetooth traffic that has been spread around right here.

So this is a great view to have to get an overall picture of what this transmission has done. In addition to that, you can also get metrics from these simulations that enable you to plot things like throughput that has-- throughput in terms of MBPS. You can also plot packet loss ratio and latency in this network in the presence of other nodes. Now, like I said, we ran the simulation with PTA. You can also run the same thing without the packet traffic arbitrator included in it.

So that about brings us to the end of this presentation, and I want you to walk away from here with a couple of key takeaways, the first one being interference is inevitable. We live in an environment where we have multiple networks that coexist in the same environment, be it coexist in time or in frequency, and you have to be able to model and simulate these scenarios in order to better understand these situations and be able to have mitigating measures so that you can have better performance.

You've got the WLAN and Bluetooth Toolboxes that provide end-to-end solutions both at the PHY and the system level for modeling these scenarios. And mitigation is possible, and you've seen a couple of different measures, including one of the example where we had a packet traffic arbitration. Or for those noncollaborative coexistence, we've talked about adaptive frequency hopping.

And so these mitigation is possible, and they can be modeled in the MATLAB environment. And finally, MATLAB provides that flexible solution. Like I pointed out, it's an open, editable MATLAB code. You can do scenario modeling. You can bring in your specific scenarios in terms of indoor or outdoor environments. You can model those scenarios in MATLAB.

And you've got plenty of prebuilt examples that you can take and use as starting points. Or in some cases, there have been you know fully established examples that you can just plug in your own parameters and model those specific scenarios. Thank you for joining me today and staying with me during this entire presentation, and I'll stay around and answer any of the questions that you have now. Thank you.