Judy Warner (00:01.532) Hi, Murthy. Thanks so much for joining us today. I'm really looking forward to learning from you today and sharing all your knowledge and insights on wireless technology. Before we get started, can you please introduce yourself? Give us a little bit of insight on your background and then what your role is at Keysight Technologies. Murali S Murthy Upmaka (00:21.26) Well, so let's start from today and then go back. Today, I am a solutions engineer and we have in our case site, we do a lot of measurements as well as we do software that helps the people to design electronic systems. And I'm more focused on system level simulation and modeling. We do have portfolio of our products that help design circuits, model the devices, and then even we are expanding into quantum and then all those frontiers that are coming up. But mostly my focus is some system level, high system level. I did work on circuits in the past. That's where I started, but I moved into the systems in the past 10 years, 10, 15 years more focused on this. And that's where I am. So before that, education, by education, I have a PhD in physics. And then I had the opportunity to take a little bit sabbatical from my first job to go to and then go and then do postdoctoral research. It's interesting. I did my postdoctoral in terahertz. That is going to be 6G in now. So it's a little bit linkage. Judy Warner (01:44.872) That's crazy. Murali S Murthy Upmaka (01:48.93) with what I did. other than that, I've been working in, joined in HP, Hewlett Packard, moved into Agilent, and then now we are Keysight. That's my background. Judy Warner (02:04.008) So in regards to wireless technology, I have a little like four or five years stint where I was working on the board level RF technologies, which to me were tough enough, much less all the physics and things that are below the surface and at the device level. So let's talk about... what's involved at the system level and then we'll drill down a bit more into my world, let's say. So let's talk about, know, we've gone, we're going from 5G to 6G and I know from conversations that I've had either in the industry or with engineers, these are hard things to perform and to get off the ground. And so let's talk about it from your perspective as a systems engineer. Murali S Murthy Upmaka (02:35.086) Mm-hmm. Judy Warner (02:57.414) What are some of the great and cool applications you're seeing? What are the challenges in our way? And just where are we with 5G and then looking ahead to 6G? Murali S Murthy Upmaka (03:10.284) Yeah, you know the the 5G in general, the communication is is something that we cannot just fathom what it was like 30 years ago versus what we have today. I mean, it's it's unimaginable. I can lift my phone and then talk to anybody on the other side of the globe, right? Or anywhere on the globe as long as they have the connectivity and then I can transmit videos. I can. Judy Warner (03:22.568) you Murali S Murthy Upmaka (03:39.768) communicate to them, I can share files, know, the, then if I can sit at my home and then do my job, that's what I've been doing for the past 25 years. And then most of the times I can move out of my office and then still be conducting the job because of the wireless technologies. You know, it's, it's, it's very cool. And then even for executives, the cool thing is they can approve on their phone, the purchase order or anything. It's a legal document, we can do it on wireless because of the security it provides. So there are a lot of cool things that are happening for the business side. But for the people that have never experienced, they don't have to know anything about the wireless, but they can still post their videos and the WhatsApp and then TikTok and then Instagram. It's unbelievable. how much the technology has come in. What is amazing to me is not the coolness of this, but what is the reach it really reached the common people in the world. Yeah, I come from India, by the way, basically from, I was born in India, but I've been a US citizen for a long time. When I go there, Judy Warner (04:49.147) Right. Murali S Murthy Upmaka (05:01.343) I see people, common, very ordinary people that probably can't even, you know, I can't think that they can afford a phone, but you know, they do have a small handset in their hands and then they're able to communicate. And then, you know, the amount of services that it has enabled them, the economies can change because of the wireless technology. That's the coolness I see the reach to the common people and then changing the lives of the people. Judy Warner (05:25.457) You bet. Murali S Murthy Upmaka (05:31.682) That's the coolness. Of course, on the higher end side, it's so cool that with this technology, what we can do to begin with, of course, we are still to work on that. That's why we'll talk a little bit about the 6G promises. The 5G promised, yeah, you can download in 10 seconds the movie, entire movie, and then you can watch. Or you can play games. Of course, that is happening. Judy Warner (05:55.374) Right. Murali S Murthy Upmaka (06:00.789) And then the other promises it made were like doctors can do remote surgeries and then, you know, and then the cars can do autonomous driving because of the 5G technology. There were so many very cool promises were made. And then in fact, that's even it's going to be possible. Of course, any technology has its hiccups. Of course, 5G has hiccups even after 10 years. That's a different story. But but. Judy Warner (06:26.385) Ha ha ha ha! Murali S Murthy Upmaka (06:29.367) But a lot of promises are fulfilled. Nowadays, compared to what I could do five, six years ago, now my download speeds on my phone are certainly much more. I can press a button. can certainly open up a 10 megabyte PDF on my phone. I can read it or I can watch a video. The experience is better, much, much better. But has it been? fulfilling all the promises it made. Looks like no. It's like you set an expectation, but you're not meeting it, right? And then the other people, like the user experience here is the biggest thing, user and industry, both. So the user experience is, what? The 5G promised me that I can just push a button and then I can just get this instantaneously on my phone. It is not happening. It is not happening. We'll dwell into that as well. Judy Warner (07:05.702) Right. Judy Warner (07:14.139) Right. Murali S Murthy Upmaka (07:27.309) discussion today, but there are a lot of things were fulfilled in 5G, but there are many things that have not been fulfilled. So it's a mixed bag of promises and fulfilled promises and not fulfilled promises in 5G world. Judy Warner (07:44.402) Well, and of course there's the military, there's radar, there's satellites in space. I mean, it's such a far reaching thing, but we don't always, as consumers, we don't always stop and think about what it takes to deploy these technologies. And of course our audience are engineers who both use this technology and they're being tasked with deploying it, which is a whole different perspective and why I wanted to talk to you about these today. So the length of development in your mind, let's talk about the trends in development and maybe what some of the obstacles are and the opportunities. And I always like giving our audience, which you've started out beautifully, which is sort of giving us a state of 5G, right? And then, and then peek over. the hill a bit and see where 6G is going. So let's talk about that first. Like what are the shifts and the trends and developments of this technology? And let's also talk about ways that Keysight and almost every other company I talked to on this podcast and beyond is talking about shift left or the digital twin. So, all right, let's talk about that. That's a handful I just gave you. Murali S Murthy Upmaka (08:47.329) Mm-hmm. Mm-hmm. Murali S Murthy Upmaka (09:05.671) Yeah. Yeah. That's a big friend. Thank you for bringing that term. left. What do you mean by shift left, right? So fortunately, there is a research paper from Air Force Research Labs some time ago. they said that, imagine there is a X-axis where, as you progress on the X-axis, the cost of corrections in your system keep increasing. on the y-axis, is the how much you know about your system, your confidence in your system as you're developing. So there are stages along the x-axis. We call them concept. You start with a concept in your mind, and then you start some calculations. We call them modeling and simulation, but still people use Excel spreadsheets. That's why I can even call it calculation. And then the third stage is you put Judy Warner (09:59.07) Mm-hmm. Murali S Murthy Upmaka (10:07.383) put the things into your prototype. And next thing is to do a limited production and then final production, then deploy it into the field. So these are all the stages that we are any product, any system product has to go through. And what they found is your confidence in the system, in the legacy way of doing things, your confidence will exponentially increase, very steeply increase once you get into the prototype and then you actually build, you go to the lab, you test it, your confidence suddenly is steeply increasing and then it will go even further as you start into the manufacturing and your confidence of being able to produce at the required perimeter of the cost and then deploying it. So what we have seen in the past is your Confidence is not so much in the beginning, not at the concept level. Maybe concept is you don't expect too much confidence, but at the simulation and modeling or your calculations are not very accurate because you're taking shortcuts. You're making too many approximations. You're putting your things into your spreadsheet and then, okay, let Excel spreadsheet calculate for it. But these are complex systems. We need to be able to do more detailed analysis before. And the previous analysis, the legacy we have doing things, let us say we found a problem at the end of the prototyping. We went to the lab, we tested, and then our confidence is high. Because of that, we found the problems in the system. yeah, it is not going to work. Right? That's very usual. And the only way you can correct is go all the way back and then do the spin. This is called in the world of electronics called spinning. you come back to another iteration, another iteration and the correction. Judy Warner (12:03.453) All right. Yeah, you spin another board and you spin another board. And during that time you're throwing out boards, chips that can't be recovered. Murthy, do you know the game, the child's game called Whack-a-Mole that they have like at Chuck E. Cheese? It's a machine where, there's a flat surface and these molds jump up and you have a little bopper and you try to hit the mole. Murali S Murthy Upmaka (12:11.052) Yeah. Murali S Murthy Upmaka (12:34.049) OK, keep hitting them. Yeah. Yeah. Judy Warner (12:36.413) And then he pops up here and you hit that one. I say, this is the whack-a-mole thing that we do, right? We, we then engineers guests like, think it's this situation. Yeah, yeah, yeah. But without modeling and simulation where you've correlated the math and you're trying to get your simulation and the physical product, it's like playing whack-a-mole, which is why we have to go into three or four spins, right? Because, um, Murali S Murthy Upmaka (12:43.1) Yeah. Murali S Murthy Upmaka (12:49.655) Let me fix it. Yeah. Murali S Murthy Upmaka (13:04.171) Yeah. Yeah. Judy Warner (13:06.297) And I always like my friend Eric Bogaten always says, hope cannot be part of the design workflow. So I think you're making this case. You're starting to make this case anyways. Murali S Murthy Upmaka (13:14.199) Right, right. Yes, absolutely, absolutely. You're right. So what does it mean by shifting left? What are we shifting? We are shifting that steep confidence increase to the left. How does it happen? If only you can model it, if only you can accurately simulate it, you know it already before you go and then put anything on the board, in the lab, in the form of hardware. Judy Warner (13:30.108) Yes. Murali S Murthy Upmaka (13:44.161) buying something, getting fabricated something in the PCB house or getting some chips to do that. Suppose you know everything, you model. Judy Warner (13:52.253) Hold on. Judy Warner (13:58.014) Sorry about that. Let me grab something. Murali S Murthy Upmaka (13:59.426) No problem. Judy Warner (14:17.319) I was trying to get to a pause point. Judy Warner (14:28.573) Go back to the point where you said suppose you're. building the board or just start there and I'll edit all this. Murali S Murthy Upmaka (14:38.017) Right, right. Yeah, yeah, I was talking about the the shift left, right? What do you mean by shift? What are you shifting? You're trying to shift that steep increase in confidence area to the left. It happens much later in the legacy way of designing systems. So by moving it to the left, you are moving that all that into the simulation calculation time, you know what we call. And so if you can model accurately, if you can model Judy Warner (14:42.928) Right. Murali S Murthy Upmaka (15:08.205) if you can simulate accurately, you know about your system already without even committing a penny on buying any hardware, right? So if you know that, then even if you have to take some correction, I mean, of course, simulating in the computer doesn't come free, right? It has some costs associated, but not as much as to the later stages. So you can correct it quickly, easily, and then your confidence level is so high that you go and then... Judy Warner (15:30.106) Uh-uh. Murali S Murthy Upmaka (15:38.008) do the prototype and then finally, and then. So this is the trend that is happening, right? So everybody wants to shift left, calculate more. You know the phrase we all know, measure twice and cut once, right? If you keep cutting, you need to bring a new one to keep cutting. Don't cut it, keep, just measure, keep measuring, keep simulating, understand your system very well before you can indulge in committing and then go to the. Judy Warner (15:51.133) Yeah. Right. Right. Murali S Murthy Upmaka (16:08.405) Measuring and making measurements in the lab is not going to be, is the video frozen? Little bit. Going into the lab, okay. Murali S Murthy Upmaka (16:29.581) Right, okay. Okay, so it is going and testing in the lab is not going to be eliminated. So the thinking for the engineers so far, anyway, I have to go in and test. Why do I need to simulate? Let me go and then do the testing and correcting. That is a hidden, there is something that is hidden that we are not realizing. Probably, hopefully, the industry is realizing slowly that there are hidden costs. there is a hidden inefficiency the way that things are done. everybody is trying to increase the efficiency of developing the systems. Shift left is a big concept going there. On top of it, when you shift left, the byproduct is you have a model, you have a system that you are building in the computer. That is what we call digital twin. So what people are doing even there, Judy Warner (17:23.695) Mm hmm. Murali S Murthy Upmaka (17:27.723) Some companies, some people, some groups are building good models in the computer, but once they go into the lab, they forget about this. Just leave it out and then proceed with testing and manufacturing and then operations, everything. No, don't stop there. You have a beautiful model. We can call it as a digital twin. Now use the digital twin to improve it through the stages of prototyping, through the stages of manufacturing, through the stages of deploying even after you deploy there is a lot of learning so please bring that back into your digital twin make it such a wonderful accurate model it behaves almost like the system that's that's out there I want to share you know the digital twin the the best digital twin the most expensive I can tell you was developed by Northrop Grumman for the James Webb telescope right when they launched yeah Judy Warner (18:24.301) I bet. Murali S Murthy Upmaka (18:26.529) They launched three years ago, was project overruns and then it took 10 years more or even more. But once they launched, it was such a beautiful thing. They predicted everything, how it should go. then, at one, this was on a beautiful January Christmas time, right? No, they launched two or three years ago, I believe. And once the satellite goes, once the rocket goes and then it start dipping. I thought, why the heck it is dipping? They said that is by plan. By dipping there, they are going to save the fuel on the satellite, which is very valuable. So look at this, modeled and then it just imitated what they did. So when they dipped, they were happy that, you know, it's not falling. It is actually by design, it went down and then it saved a ton of fuel. So now we have this. wonderful James Webb telescope in the space. And then we are seeing a lot of things that it's improving our understanding of physics, et cetera. Anyway, that's a different story. I don't want to get excited about it. Judy Warner (19:32.209) Well, and that one, if anybody was paying attention to the James Webb telescope launch, which I was, and I bet you a lot of our geeky audience was, the stakes were so high. Like, because if it blew up or something went tragically wrong, the time it would take to redeploy was unfathomable. Murali S Murthy Upmaka (19:42.924) Mm-hmm. Judy Warner (19:57.667) the stakes were so high, it's like they had one chance and it had to be perfect, which is why it took 10 years, but it could put them back another 10 or something. So it really was a marvel to behold that they were able and we're talking thermal power, fuel efficiency. I mean, so many things in that system to model. So it really is the ideal thing to look at and Murali S Murthy Upmaka (20:19.784) Exactly. Judy Warner (20:27.832) against the complexity of the things we're say for radar or cell phones. And so you'll like this because on my side, Murthy, on the board side, RF engineers would come to me and say, why aren't the simulations I'm doing matching? The boards you're giving me. I'm like, why they? Murali S Murthy Upmaka (20:32.18) Absolutely. Judy Warner (20:54.65) The things that I have to tell you from the board side that we can do that's going to screw up your performance number, but they're blind to you. So you have to make sure you put the board effects in there in your model too. And so, you know, I've seen pieces of this. and it really is true that I wrote a blog once that the title was slow down to speed up. And that's what I hear you saying. It's going to cost you a little bit of money. It's going to slow you down at first, but long-term. Murali S Murthy Upmaka (20:54.87) you Murali S Murthy Upmaka (21:04.503) Yes. Murali S Murthy Upmaka (21:16.742) yeah, yeah. Judy Warner (21:24.614) Like you said, your confidence curve is going to go straight up and your time to market is going to get better. But it's not, we haven't baked this into our psyche, I think is an industry yet, but we're getting there, aren't we? Murali S Murthy Upmaka (21:41.892) yeah, absolutely. The system level problems unfortunately are so expensive to address than the problems in the small circuits. mean, circuits can create problems. I'll give you another example that I came across in my career. So this company was putting all these towers in the communication towers, right? So they have these amplifiers inside and They didn't simulate much, know, they just, somebody designed and then, you know, they said, okay, it is working in the lab. And then they took it there and then they put it there. And then each amplifier started what we call thermal overrun because of the proper biasing, certain changes. And they said, okay, we are going to make it corrective. You know, we are going to correct it. So they already installed say 500 installations. Just this $5 amplifier to replace Judy Warner (22:27.516) Mmm. Judy Warner (22:40.219) gosh. Murali S Murthy Upmaka (22:41.747) on each one of those towers, costed them $10,000. It's expensive. You know, had to send a person, go in and replace it, and then, you know, again, test it, that it works there. It's going to be expensive. So the system level problems need to be addressed upfront. Otherwise, it's going to be really nightmare for designing. That's why I like this concept of digital engineering, the digital twin that the concept has come in. We have to shift as much as possible to the left before we go and then do the test. Judy Warner (23:24.174) And then we have that single source of truth that lives on, right? And if another engineer works on it or whatever, whatever, you have all that record and you have all that good rich data that you can keep so someone doesn't have to reinvent that painful wheel again, you have this, this single source of truth, which is a wonderful thing. Murali S Murthy Upmaka (23:24.823) That's the. Murali S Murthy Upmaka (23:29.94) Exactly, exactly. Murali S Murthy Upmaka (23:40.309) Yes. Murali S Murthy Upmaka (23:48.63) Right, right. Right, right. I just want to add one point though here. Maybe we can come back to it. You rightly said about the circuit board and then not taking into account the parasitic effects of the layout and things like that. Even after you take everything, let's say you design, nowadays integrated circuits are designed to such an accuracy after you go and fabricate invariably. Judy Warner (23:51.972) All right, well, we're going to talk more now. Go ahead. Murali S Murthy Upmaka (24:19.533) 99%, maybe 90%, they work the way they are simulated, right? Because of the simulation tools that are out there in the market are so accurate. So circuit level, we are seeing good progress. You can predict, you can get the same answer when you measure. But the issue is you can't use, now let us say you have 10 circuits, you have to put them together. Each circuit takes finite amount of time to simulate. That is OK while designing that single circuit, but when you put 10 of them and then you have to run through thousands or even hundreds of thousands of simulations, the simulation time goes through a roof. So the only solution, the solution for that is convert your circuit level. You should be able to capture the behavioral model of your circuit and then put that and then integrate. Now this behavioral model. Judy Warner (24:53.931) Yeah. Judy Warner (25:01.946) I bet. Judy Warner (25:13.244) Hmm. Murali S Murthy Upmaka (25:17.653) Modeling is a big thing in Keysight technologies. That's what I like about it. The tools that I use and then, you know, that's what I share with my designers. Please spend a little, slow down a little bit and then create a good model. Create a good model, behavioral model. And then if the model can come from measurements, that's what Keysight does, lot of measurements. Bring those measurements, model it, and then you have a much better. Judy Warner (25:21.167) Okay. Judy Warner (25:30.382) Yes, slow down to speed up and save money. Yeah. Murali S Murthy Upmaka (25:46.306) picture and then you when you do the completed overall system analysis, your simulation time is manageable. I will not say it is lower because now you you're able to simulate. You can see a lot more statistically whether the system is going to function or not. So there is that trend is also going on behaviorally modeling and. Judy Warner (26:07.204) No, it's good that you brought me back to that simulation of the system. I interviewed Anurag Bhargava, your colleague, and we talked a lot about the heterogeneous designs and the challenges. In his case, he was working on RF modules and stacking those technologies and getting them to play well together is a real challenge and more and more I think engineers are going to have to face. Murali S Murthy Upmaka (26:16.801) Yeah, yeah, my colleague. Murali S Murthy Upmaka (26:21.92) Yes. Judy Warner (26:37.18) Maybe what they learned in college and this and that, but the complexity of systems is going to force us into thinking about heterogeneous integration. And so I appreciate you bringing that back to the forefront. Okay, now let's shift to the 5G. You taught me something and maybe our audience knows because I'm always learning, but can you help me understand where we are with the 5G TN? And then you've taught me 5G NTN. So can you start by defining the first one and then we'll jump into the other one for sake of this conversation. Murali S Murthy Upmaka (27:15.765) Right, so anything on the terrestrial, right? know, of course, something traveling in the train is still terrestrial. It is moving, but it is all the base stations are stationary in a terrestrial network. And then the user equipments, they can be moving, they can be stationary, they can be computers, can be iPads, they can be phones, they can be anything, know, user equipment. They can be doorbells and then, you know, Judy Warner (27:44.548) Yeah, anything on earth. Murali S Murthy Upmaka (27:44.746) video, IOTs, anything on the earth. So that is all the terrestrial thing that we are talking. And then the communication between all these devices is orchestrated, monitored by the network. That's what we call the terrestrial network. So what happens is in all these systems that I deal with, invariably there is something that radiates, we call transmitter, and something that receives, we call the user. equipment receiver. So, and in between, of course, the space that's hostile to helping these two talk to each other. We'll come to that as well. But the network, what it does is it's not just one base station talking to one phone, right? It's one base station talking to hundred phones and there are hundred base stations. And then which base station as the phones move, each base station should be able to hand over to the next one. So all these operations have to be orchestrated by some intelligence we call network. is it the network? is the, yes. Right. So the network that controls, that helps the communication between all the terrestrially located devices is terrestrial network. Now what happened is in the past, Judy Warner (28:51.974) So that is the TN, the terrestrial network, 5G terrestrial network is everything you just described. Murali S Murthy Upmaka (29:12.973) 10, 15 years where we are seeing launching things into the space has become much cheaper and easier. then, you know, the technology has improved. The cost per kilogram to launch into the space has come down dramatically. So that now people started thinking, OK, we need another one, another promise for 5G. You know, I said one of the unfulfilled is ubiquity. The promise was. we will make sure everybody is able to talk everywhere, to everything, anywhere, anything. But that was not really possible with 5G because of the costs and then certain things that came in the way. So suddenly now people started thinking, why don't we bring the satellite communication technology into this fold of the wireless? There were already GPS phones, people had Judy Warner (30:09.904) Mm-hmm. Murali S Murthy Upmaka (30:10.849) Very expensive, right? If you want to make a call, probably $10 per minute or whatever it is. But if you want to really establish the ubiquity, you need to bring in new technology. So they thought they will exploit this non-terrestrial objects like satellites. It can be a drone. It can be a plane. Anything out into the space. Judy Warner (30:33.286) Mm-hmm. Murali S Murthy Upmaka (30:38.871) But again, somebody has to manage all those devices as well. That is the non-terrestrial network, NTN. The problem is there are two networks now, like two intelligent people controlling two different systems, but they have to hand over to each other. That is the handing over a call from terrestrial network and non-terrestrial network. That's what is broadly the NTN. Judy Warner (30:46.19) Okay. Judy Warner (31:00.112) Okay. Judy Warner (31:03.75) Got it. Judy Warner (31:08.576) I think you have a slide you can throw up here, which I'm not, we don't have to record this part, but I think what we'll do is we'll show that slide that shows the terrestrial and the non-terrestrial there. Okay. I'm making a note for myself. Murali S Murthy Upmaka (31:23.978) Mm-hmm. Mm-hmm. Judy Warner (31:31.344) Okay. Judy Warner (31:37.336) Okay, now I'm gonna respond to what you just said. Let me get another drink. Judy Warner (31:56.474) Okay. Judy Warner (32:03.686) Well, Murthy, when I think about that handoff you just mentioned, you know, the terrestrial and the non-terrestrial bandwidth. How do you get those to communicate and what are the challenges say to the engineers and our audience that may be developing? We've been talking about IoT for years and years and years and almost everything has wireless in it now. So what are the challenges specifically for our audience and also bandwidth issues and how that applies to the digital twin simulation you've been mentioning. Let's put those two ideas together. Murali S Murthy Upmaka (32:50.861) So the the challenge is, of course, you know, it's like to, as I said, if you put two people together, they will never be able to communicate. They conflict each other. And then I want to do the things my own way. then, you know, exactly. But but you know how to bring in the coherence between them is a big thing. It's a lot of. Judy Warner (33:03.834) Hehehehehe Judy Warner (33:09.308) It's like putting two engineers in a room. Yeah. Murali S Murthy Upmaka (33:20.101) You don't want to modify the terrestrial network already designed and then there and then you your non terrestrial network software you're writing in it to be very compatible with this terrestrial. So it's more in the software realm, but it also brings into the physical world as well as the challenges, right? know, so the designing the link, what we call, for example, one One way is I want to communicate, I want to talk to somebody somewhere, maybe sitting in Himalayas. Let us just simply say that. I can make a call, it goes through my base station nearby and then it connects to the satellite and then the person in there in the remote area can, it's what we call direct to phone, receives the signal from the satellite. So this is a lot to ask for. There is so much going on here, the signals can be very weak, you know, because you don't have, in the phone, you don't have too much power to transmit to the satellite. And then also the satellite to the ground is a lot of loss, tremendous amount of loss. So we have to model all these things. And then in addition to that, when the satellites fly, Judy Warner (34:36.419) Mm-hmm. Murali S Murthy Upmaka (34:45.057) that there are other effects and it's not just the loss. It is because of the motion that translates into some complexity in the signals. We call them as Doppler effects. And then the delay, we had to take all this. We had to synchronize the ton of things for the call to be mature. And let's say we are communicating and unfortunately the area that the person remotely he was driving, let's say, and then he gets into an area where there is a base station. So we don't want this link to be again going through the satellite. How do we move that call from this non-terrestrial link to the terrestrial link? Because now suddenly the base station can, we have this beautiful terrestrial network in the world, right? And when I do this, it goes to the base station and the fiber and then the fiber through the ground and then through the ocean can go there, connect to another base station. So there is a lot that happens in handing over. So this also should work flawlessly. Everything, both the networks in terms of their intelligent communication, accepting what the other network is telling the network. And then in the real world, the signals, their losses, and then their quality, all these things accumulate. It's a lot of challenge. And then, of course, I'm not saying that I have knowledge of everything that goes on there, but what I know is electronics, part of this, is what we deal with in Keysight and then what we deal with in our simulation modeling tools that we give to the industry. We design the electronics in here, and then this electronics is so... you have to model it so accurately in order for you to be able to predict. then modeling accurately, I always go back there building the behavioral models and then simulating them. then another important thing is this channel between them is so dynamic. I tell you that the dynamism of this channel is so important. Let's say you're holding your phone, you're talking and then Judy Warner (37:01.615) Mm. Murali S Murthy Upmaka (37:10.913) the base station is communicating to you. And then we have some advanced concepts in 5G, what we call beamforming. it can actually, previously in the older wireless technology, the base station was transmitting everywhere. And then, okay user, if you can pick up my signal, pick it up and then go do whatever you want. That was the attitude, but the base stations have changed. They now say, okay, I'm sending my energy directly to that user. the information the user has to receive is so pointed towards the user. It's much more gain and then much more less interference because of that. So the user is able to receive very easily. Now think of the user is holding the phone. So the user is, the beam is coming and then pointing just turns his head, turns around. Everything is gone. You know, the... The whole process has to start and then we have to again make sure the base station and the user are communicating through these narrow beams. This is one of the biggest things to highlight in 5G world. The challenge is they thought, okay, we can do it in millimeter waves, 28 gigahertz. We have a lot of antennas, we can put them on the... on the phones and then on the base station and then we can do all this, but they realized this beam management, what we call one of the biggest things, is not easy. that's what, anyway, that's what they're trying to tackle and then they're going to tackle that a little bit more intelligently into 6G. That's what is happening. So these are some of the challenges. The procedures, the way we manage the things to counter what the space in between the transmitter to the receiver happens. And then the environment, so many devices being there, all these things were problematic. Again, the thing in the beginning I said, the technology is reaching the common man. Only then it will be successful. Only then the cost of being able to operate Murali S Murthy Upmaka (39:28.749) will come down and then only then the companies can make money, et cetera. You won't believe how much of course in USA for wireless plan, typically you pay 50, 60, maybe 80, $100 depending on the provider. It's not that much in those countries. They pay maybe one tenth of what we pay, but they still get the same quality because of the massive adoption by the population. the companies can still make money. So that is the important thing. So for us, some of the things that 5G with all this challenge, this 5G, if it fell short, it fell short because of the adoption, know, the return on investment for these companies. So what they have to do is how do we make this more technologically acceptable in the sense, how can we bring the cost till I can make money? That's what if AT &T, Verizon, and then all these people think. They have to find new use cases. So as you mentioned, the IoT, that's a new use case. What if we provide a service that monitors these devices everywhere, especially the critical infrastructure, the mining, the their shipping, know, you're shipping these things. So through the NTN, if you can monitor all of them real time and you can make money, then of course you can invest that into the other side and then make the promises of 5G more thoroughly fulfilled. So the IoT is the most important thing. And then what it offers, Terrestrial IoT we do have, of course, all of us doorbells and then everything we have cameras and then everything that is becoming more and more common compared to what it was 10 years ago. But the critical infrastructure and then offering those services will bring more money because it's an industrial use case for these companies and that is more reliable. What happens if there is a hurricane, if there is a twister, suddenly Judy Warner (41:44.997) Right. Murali S Murthy Upmaka (41:46.466) The terrestrial thing can be broken, but from the satellite, can still beam, and then you can still receive the signal, and then you can still be reliable and resilient. So this is the important thing that they are trying to do. So balance the use cases, make money in one area, use the same infrastructure to cater for the other before it matures. That's what is happening. Judy Warner (42:11.353) Well, I think maybe because of COVID and all the disruptions there, the idea of countries protecting their critical infrastructure became a big topic that anybody can hear in the news on a fairly regular cadence that this is disastrous because we are so dependent on the terrestrial network now. And then there's geopolitical things and Murali S Murthy Upmaka (42:23.703) Mm-hmm. Yeah. Murali S Murthy Upmaka (42:29.175) Yes. Yeah. Judy Warner (42:39.867) on and on and on. you know, people need, you know, heat and water and communications and all these things in an uninterrupted way because there could be loss of life. So I think that's certainly driving it. And when you're talking about keeping those costs down, is your thesis here then that by shifting left, it helps bring those costs down? Murali S Murthy Upmaka (43:09.257) Exactly. Judy Warner (43:09.303) which allows sort of a buy-in across and to scale and to be affordable to companies that are investing, you know, making these large investments. Murali S Murthy Upmaka (43:22.333) Exactly, you know, we have been in a constant pressure to deliver the, you know, the systems are becoming more complex. They're becoming larger systems. And then, but still the end users expect they don't want to pay more. The industry doesn't want to pay more, but these, have to do that. So you're under constant pressure to deliver in time. You know, the time to market is so important. So there are so many pressure points on the engineers. That's why I won't blame them if they take shortcuts. I want to go to the lab as soon as possible and then test it out. And then anyway, I have to test, I'm going to fix the problems there. That mentality is okay, but it's not for the modern era of developing systems because systems are more complex. You can't take the same approach and then... Judy Warner (43:52.271) Mm-hmm. Murali S Murthy Upmaka (44:17.559) build the inherent inefficiencies and then think that, okay, this is the cost. No, that is not necessarily, that has to be the cost. You bring it, shift left, see how much you can save. All that you have to do is apply to one project and then reap the benefits and then you will see going far into the future, the benefits of, and also I would not just underestimate Judy Warner (44:26.297) Mm-hmm. Murali S Murthy Upmaka (44:47.181) Even in simulation and modeling, is a by and large, there is a part of science that is a part of a little bit of art as well. That's your knowledge, your experience. The way you set up, the way you run, you can get things done faster. You can get things done more accurately by a little bit common sense, even in the PCB world. I can give you an example that because you brought PCB. So all these electromagnetic tools, they give you very accurate results. Judy Warner (44:56.379) Mm-hmm. Murali S Murthy Upmaka (45:16.397) because you're really going into the physics and simulating what happens inside. But if you're designing a filter, and then depending on the tool you use, the central frequency of the filter can be underestimated or overestimated. Based on the tool, the technique you use, whether it is a momentum, maybe Anurag might have mentioned these things, it can be higher and lower. But an experienced person, I have seen this literally with lot of users there. Judy Warner (45:20.708) Mm-hmm. Murali S Murthy Upmaka (45:45.994) yeah, know how much it is going to, the tool is going to shift. So I will just design for a slightly different frequency so that when I measure it is going to be at the frequency I want. It's a very tiny fraction, but very clever. And then that is the art they learn. So there is a little bit of art you have to apply as well for simulation modeling to get what you want. But once you master that, your costs are going to come down unbelievably, you know, controlled. Judy Warner (45:55.515) All right. Judy Warner (46:06.576) Right. Murali S Murthy Upmaka (46:14.135) You don't have to toil too much in the testing world initially. Of course, final test is nobody can eliminate. We should not be avoiding any testing. It has to be done thoroughly. Only then you deploy. designing it upfront, is what shifting to the left is so important for all these systems. Judy Warner (46:39.737) Well, if in our audience, if there's any of you engineering managers out there, here's a gem for you. Make that case and that fight because often the engineers that are designing circuits or designing the circuit board or just looking at the signal integrity, you know, we get into these silos because of the complexity. But if we don't have that. Murali S Murthy Upmaka (47:01.271) Mm-hmm. Judy Warner (47:06.523) One, those systems in place across the system, then the whole thing's broken. So, you again, you invest wisely to save and scale. So there is a win. It's just, I say I always have engineering empathy, Murthy, and I bet you have a lot of engineering empathy too, because the pressures on engineers today is insane. Murali S Murthy Upmaka (47:11.383) Yes. Murali S Murthy Upmaka (47:29.773) Right. yes. Yes. Judy Warner (47:35.142) to save money, to do it faster, on and on and on. But I've been beating this drum way, way, way downstream at just the little circuit board level of how much time and money can be saved. But that's true all the way back to the concept and the device, all the way through the system. So engineering managers, think about this one. I'm sure you have tools and all that, but I think things are changing and we have to pay attention. especially with critical things like 5G. Okay, now for a fun question before we wrap and thank you so much for your time. This has been such an interesting conversation. Okay, we're having trouble, you know, getting 5G and getting the terrestrial and the non-terrestrial to communicate and line of sight and beamforming and all these things we've talked about. Murali S Murthy Upmaka (48:11.255) Mm-hmm. Judy Warner (48:32.879) But you hear it all the time, Murthy, them talking about 6G. So what's coming? What do you see this ahead? Because I always like to ask this question of Keysight people, because you guys are always way out there on the leading or bleeding edge, and often sit on standards committees and you see this ahead of the public. So I'd be very interested in you sharing with the audience what you're seeing. Murali S Murthy Upmaka (48:55.309) Mm-hmm. Murali S Murthy Upmaka (49:00.535) Sure, sure. You know, what comes to my mind? There may be lot of ways to answer your question, which is many different angles, but the angle that I take is 6G, the first thing that is going to happen is these two networks that I mentioned, NTN and terrestrial and non-terrestrial networks, they are going to be not anymore two separate entities. In 6G, they are designed to be one entity. Judy Warner (49:28.507) Mm. Murali S Murthy Upmaka (49:29.769) So you buy design, so it's in one intelligent head. There are not two heads now, one thing. So that makes, I believe for the people designing and developing, it's almost like going back to the drawing board and then designing the whole thing, right? So that is 6G. So they're going to merge 5G and NTN and TN and then it's going to be one single network in 6G. Judy Warner (49:36.633) Okay. Murali S Murthy Upmaka (49:59.864) for sure is the biggest thing that will happen. And then that opens up a tremendous amount of opportunities and use cases. Some of the use cases, they're very interesting application. Like 5G, we said remote surgery and then the other things we said. But here 6G, it may bring a little bit more interesting stuff because As the frequencies, 6G is going to happen very higher frequencies, a lot higher than what we are doing today. And then the higher the frequency, one big thing, one benefit we get immediately is most likely it is unused frequency. So for the government, they don't have to find new spectra. They can sell. then possibly all governments together can sell the same spectrum to these people because You know, the 5G implementation in the USA is a different frequency bands, slightly different from what is in Japan, what is in Europe, and what is in China. So everywhere the systems are different, but probably they can give the same frequency and same frequency bands that will unify development all systems for that particular band. know, so there are not too many things. That's one advantage. And the larger the bandwidth, you can send a lot more information easily. Judy Warner (50:59.483) Mm. Murali S Murthy Upmaka (51:22.657) communication theory, all of us know. But more than that, what it is going to fuel, the way they are saying, there are five or six fantastic use cases. Number one is basically, because all this Beam management that I mentioned, and then also the way the user equipment communicates to the base station, it has to send some information, what we call channel state information. Judy Warner (51:41.915) Mm-hmm. Murali S Murthy Upmaka (51:51.694) things like that for them to reliably operate. They are so dependent on so many parameters, hundreds of hundreds of parameters. That means parameters that are dynamically changing. That means your algorithms, whatever you have today, they have to operate very quickly in real time. That is the problem with 5G. We could not come up something that will manage less than a millisecond. As I said, you I turned my head everything is gone. I have to realign my beams for a reliable connection. So that is what is in 6G world. Everything is going to be based on AI ML. Artificial intelligent machine learning everything literally. So that is the fundamental shift. So when you say artificial intelligence machine learning, what is going on here is. By nature. Judy Warner (52:38.532) Wow. Murali S Murthy Upmaka (52:49.911) the way neural networks are designed, they can handle hundreds, thousands, maybe hundreds of thousands. Now we are hearing millions of parameters easily. They can absorb, they can react, and then they can predict what needs to be done. So that could be the solution for these intractable problems today. So that is going to happen. Somewhat it is coming in 5G, but that is fundamentally, 6G is based on AIML. Let's come to the use cases, very interesting use cases. As I said, autonomous driving, right? You're driving in a car. Today, if you are in San Francisco, very interesting, the Waymo, you can order and then it comes and picks you up. But the what is that Waymo car is able to see is what is visible to it. When I say visible, it can be optical, it can be microwaves, but everything that is visible, something is behind the building. Judy Warner (53:39.96) Right. Murali S Murthy Upmaka (53:47.382) It cannot see somebody walking through the street and suddenly comes. It cannot predict that somebody is coming. So there are many, many things that are going to be missing, whereas in the 60 world, everything is connected to everything. And then there are so many sensors that are communicating, not the sensors on the car actually acting alone on their own, but these sensors are communicating and the network will have a. Judy Warner (53:49.912) Yeah. Yes. Yep. Murali S Murthy Upmaka (54:16.631) picture of what is the surrounding of the car, not just in the visible region, but beyond that. So once the network knows, it knows how to drive the car much more intelligently. If somebody is coming, it will say, okay, stop or slow down, or somebody is coming, it can caution. So it is going to be more realistic. So how are we going to sense these things? It is all radar technology. It's a wonderful thing. The communications. and radar world are going to merge in 6G. And there are a lot of research that is going on, very exciting research that is going on, to bring these two technologies together in the way of creating new waveforms and then new techniques. then so the sensing income, that's what we call integrated sensing and communication, ISAC. It's a use case. When that happens, You know everything everywhere. Probably even when not just the cars, when you're walking on the street, know, something is happening, you will know what is happening in the next street. can literally, the network can inform you whatever is going on around you if you want to know. The other application is there is a transmitter, there is a receiver. That's all we are dealing with in the communication systems. So when the transmitter communicates to tries to communicate with the receiver, it transmits these waves and these waves are going to be attenuated. They can be reflecting off the buildings and then, you know, there can be other radiations coming. So what comes back into the receiver is not just what the transmitter sends. There is a lot more that comes back. And then all that interference, all those things can cause what we call deep fading. That means signal can fade. Judy Warner (56:02.554) Mm. Murali S Murthy Upmaka (56:10.091) So you may not get anything, or you may get better signal. Both things can happen. So the receiver receives the signal, and then it estimates what is going on between the transmitter to the receiver. That's what we call channel estimation. And it conveys how good the channel is to the base station. And the base station, in turn, will modify the waveform. Sometimes it'll say, I'm not going to communicate to you at the very high speed, but I'll reduce the speed because the reliability is more important than the speed, right? Always we say safety is more important than being able to do fast or drive faster. So same thing applies here. So what is going on here is the channel. Judy Warner (56:51.556) bright. Murali S Murthy Upmaka (57:06.049) we are trying to counter the channel. We are living with what the channel is. In 6G, it's total shift. What if I quote all the buildings? This is typically more problematic in the urban area, right? I go to New York or San Francisco and then quote all the buildings with electronic coatings that I can control. So if I can control those, what we call intelligent surfaces programmatically, how the beam, how it should reflect, where it should reflect and send the reflection. So if I can coordinate all the reflections, combine in such a fashion to the receiver, suddenly the receiver receives beautiful signal, wonderful high signal to noise ratio. What it means is we are taming the channel. We are creating our own channel. We are not living with what is happening. We are creators of the channel. It's like God creation, right? Judy Warner (57:59.757) Right, right. I was just... Don't you... I know because of how long you've been in the industry and how long I've been in the industry. I'm not sure a 30 year old would appreciate it in the same way. But boy, for people like you and I, it feels like science fiction unfolding, right? In real time and it always amazes me how much Murali S Murthy Upmaka (58:03.277) So that is another application. is called. Murali S Murthy Upmaka (58:22.711) Yeah, yes. Judy Warner (58:30.478) the human race and technologists and engineers have been able to do in a matter of decades. It is just truly stunning. And the pace of information or the pace of innovation is speeding up. And I'm sure it's going to be powered by these technologies you're talking about with AI, ML and gosh, it is something to behold. But then as you said, finally we're going to deliver and 6G will deliver on the ubiquity that we've been promising all along because we've eliminated the limitations of line of sight and beamforming and those things. And we now have the terrestrial and we have the means to have two intelligence systems communicate flawlessly. That's amazing. Well, Murthy. Murali S Murthy Upmaka (59:08.299) Yes, absolutely. Murali S Murthy Upmaka (59:24.971) Mm-hmm. Mm-hmm. Yeah. Judy Warner (59:29.366) I've taken so much of your time, but thank you so much for sharing your vision, your wisdom, your engineering knowledge and what is here now, what engineers can do to get better. again, lean into the shift left, the digital twin, lean into tools like Keysight. Where can our audience go? I'm going to ask you this, but I don't know that you know it. Do you know where Dolav is going to send us? He's going to send us, here it is, to a boot camp. here it is. There's a solution brief, right? Okay, can I ask you about that? Murali S Murthy Upmaka (01:00:07.689) There's a solution brief. There is a boot camp videos, hours of videos. There are wonderful tools that are out there on keysight.com. All that you have to go there and search for whatever material you want. Typically, Judy Warner (01:00:12.58) Okay. Judy Warner (01:00:24.12) Let me reframe the question and have you answer it because I feel like I missed a step there. Murthy, where can folks go to learn more about this and engineers that may want more information about the intricacies? I know you have an amazing knowledge library. Where can folks go to learn more? Murali S Murthy Upmaka (01:00:49.293) People can go to keysight.com basically, right? The search engine is wonderful if you ask for whatever you're looking for, 5G and TN, let us say. You type it and then it should automatically come. There are sometimes keysight.com slash find slash 5G and TN, as simple as that. It will find the whole web page dedicated for 5G and TN. Similarly, we have page dedicated for 6G and then, you pages dedicated for radar. So you can really jump in, get all these things. And also look for sometimes under the support, you do have learning tools, lot of learning tools. So you can also search for solution briefs or application notes, product notes. There is always a ton of things. Very specifically coming to the area that I deal with, because Keysight is a big ocean. If you go to the Keysight, there is the products and then, you know, there is the software, under software design engineering tools. And it shows the entire portfolio of our tools. And you can get in there, very specific areas. If you're doing a circuit, you can get in there. If you're doing a system, you can get there and then ask for it. Judy Warner (01:01:47.503) Mm-hmm. Murali S Murthy Upmaka (01:02:13.035) So there is, it's always easier to get there and then find. If there is any specific question there, know, we are always very happy to very quickly answer. Just enter your question under the support, contact us, or there is a place and our field is worldwide, very ready to get you all the material that you want. Judy Warner (01:02:39.95) Well, thank you again. And I will get for our audience. I will get with Murthy and pull a couple. Besides all the things that he just shared, I'll get a couple very specific to our conversation. I'll put those in the description and the show notes. So make sure you go check that out. Murthy, it's been a delight to meet you and to learn from you. And thank you for all the insights you are giving to us in this conversation, not only to the practical consumer devices that we're facing, but also how engineers can solve some of these problems and these complexities they're dealing with daily and just sharing your knowledge and wisdom. Thank you so much for joining the show today. Murali S Murthy Upmaka (01:03:23.703) Thank you, Judy. Thanks for the discussion. It's very exciting. Always a pleasure to share. Judy Warner (01:03:29.721) our audience make sure you go check out the description in the show notes I'll put all I can for you there and we will see you next week until then remember just always stay connected to the ecosystem