On behalf of Qorvo, I want to welcome everybody to our 2018 Investor Day. I want to express how pleased we are to have all of you here with us today. We've got a full agenda, and we're eager to get started. Before we get started, I want to remind everybody that the safe harbor language that applies to our press releases also applies to today's presentations. We'll begin the morning with presentations from Bob Bruggeworth, President and CEO, followed by James Klein, President of IDP, then Cees Links, General Manager of Wireless Connectivity. We'll take a 15-minute break, then we'll be followed by Eric Creviston, President of Mobile Products, Todd Gillenwater, CTO of Mobile Products, Steve Grant, Corporate Vice President of Fab Technology and Manufacturing, and Mark Murphy, our Chief Financial Officer.
We're also joined in the room by Gina Harrison, our Corporate Controller, Roger Hall, General Manager of Defense and Aerospace, Dave Luzinski, Senior Director of Strategy and Business Development for IDP, and Josh Adams in investor relations. We are scheduled to conclude the formal presentation at approximately 10:50 A.M., giving us plenty of time for Q&A until we conclude at 11:30 A.M. With that, I am very pleased to hand the podium over to Bob Bruggeworth.
Well, good morning, everyone. I want to thank you for joining us today. Those of you in the room, we appreciate you coming in. Those of you listening online, appreciate you doing that. Those of you listening to a replay, hope you enjoy the show as well. The team is certainly excited to be here today, my comments will be relatively brief. I do want to let you know that much like last year, it is extremely rewarding to be working in this industry that's connecting people, places, and things. When I think about Qorvo, I do think of us like an Internet of Things company, you're going to hear a lot about that today. Whether it's connecting people cellularly, Wi-Fi, Zigbee, Thread, et cetera, we have a place there.
I've said it before, I'll say it again today, it continues to be a great time to be in RF. We feel we have a unique position in the market. You're going to hear a lot about our market focus. I'll touch on that in a minute. Also, our premium technology portfolio and our operations excellence, we're making great progress there. If I look at the markets that we're focused on, there's seven markets, six in IDP, obviously, one is mobile. About two years ago, James and the IDP team repositioned their product portfolio to go after the highest growth segments with the products and technologies that we have available to us. Quite honestly, they've put down two straight years of over 20% growth.
We'll look for that to continue, and what James and Cees are going to talk about are how we're going to continue that growth as we look forward. In mobile, since we formed Qorvo, we've worked hard to continuously improve the technologies and our process, along with our manufacturing, to reposition our portfolio into the highest growth, most complex opportunities, and we're making great progress there as well. When I look at our technologies, real proud of this, and today, what I want you to listen for are a few key things about each one of these technologies. You're going to hear about SMR BAW from Qorvo and its power handling capabilities and its ability to handle higher frequencies. Pay special attention to that. The ability to handle higher power is one of the requirements that we see emerging inside the handset.
Not talking about power class 2, but the amount of loss after the PA that the PAs have to overcome and its impact on filters. In SOI, you're going to hear about our proprietary SOI technologies, our third generation, and some of the improvements that we've made there, and it's truly impressive. In gallium arsenide, you're going to hear about the work that we're doing to improve its performance in higher frequencies, along with putting out higher output power and improving efficiency. In gallium nitride, you're going to hear about how gallium nitride helps our customers lower their operating costs and improve their reliability. Great progress in GaN. Envelope tracking. In envelope tracking, you're going to hear more about the work that we're doing to extend battery life, along with improved thermal management.
We also believe we're leading and using envelope tracking for the broader bandwidths in 5G, and also in packaging, the work that we've done to improve our integration capabilities. From an operational excellence perspective, last year, Eric talked about our world-class customer support, particularly in our application engineers. That applies to both IDP as well as in mobile. We've improved our R&D effectiveness. We continue to scale our organization. Steve will spend time on our clean launch program. We've made tremendous progress in our ability to ramp products on time without defects. Capital efficiency. We're doing a good job of growing our top line, improving our margins, and also improving the amount of our return on our capital. You're going to hear about that today as well. We've also begun our journey on lean. Last year when I got up here, we'd really just begun our lean journey.
We're making great progress. We're very fortunate to have someone by the name of Jamie Flinchbaugh, who is a lean renowned expert, who is on my staff and is helping us modify and change our culture. The organization today uses words that quite honestly, years ago, some of us didn't know. We are holding Kaizen events. We're doing Gemba walks. We're working on A3s. I'm really proud of how the organization is embracing the lean culture changes that we're making. Very pleased with the organization. Putting it all together, we have the core enabling technologies. We have multiple long-term growth drivers. Our markets are behind us. We're targeting the highest growth opportunities and the most complex opportunities, and quite honestly, doing a very good job at our production ramps, and you're going to hear how our customers are recognizing us for that.
Before I hand the podium over to James, a few words about IDP. IDP's really, as I said earlier, come off a couple good years. This past year, they grew over 20%, gross margins improved, OpEx as a % came down. Operating margins were about 30% for the year. That three years that we've been Qorvo, they've made tremendous progress. Very pleased with that. In fact, when I was at the IDP sales conference two weeks ago, I shared with them that I thought the IDP team and James had a championship year. Also reminded James that I thought when you're a true champion, you find yourself on a Wheaties box. To be honest, I found one. Before James comes to the podium, we have a short video.
All around you, Qorvo's IDP is connecting and protecting what matters. New breakthroughs in GaN, Si, and low-power wireless technologies are creating more powerful RF and energy-efficient chips that are building the wireless infrastructure needed for a 5G world. This explosive growth enables an incredible cutting-edge connected world, including autonomous vehicles that create safer, more efficient transportation, seamless smart home connectivity, and GaN technologies for the global defense and aerospace community, creating partnerships for commercial applications that build on more than 30 years of R&D experience. Join us, and together we'll build a more connected and protected world on the ground, in the air and all around you.
As you know, Bob congratulated us for a couple of years, and of course, before he got the sentence out, he said, "I'm looking for a three-peat." About the time he got that sentence out, he said, "Well, maybe you can do like UCLA and do six times in a row." We certainly have the gauntlet set for the organization. Before I start, my view is there's really, at my level, only two things I can control. I can control the strategy of the organization, and I'm going to talk a lot about what our strategy is as we go through the presentation. I can establish a team. I think in IDP, we have established a world-class team, and I want to spend just a couple of seconds introducing the folks that are here from IDP.
Dave Luzinski up here in the front runs strategy and business development for the organization. If you get a chance, talk to Dave. He's been instrumental in setting the strategy you'll see today and also in identifying and closing some of our acquisition targets over the last couple of years. Roger Hall is the GM of our high-performance systems. He's responsible for our base station activities and also responsible for GaN and all the defense business. There are a lot of growth activities in Roger's business. Cees Links is the GM of wireless connectivity. Cees is responsible for all the Wi-Fi, Zigbee, BLE chips that you'll see, and also tremendous amount of growth. Quarter after quarter, you've heard about us growing in defense, you've heard about us growing in Wi-Fi and IoT, and these are the two guys who've been responsible for that growth.
Let me talk about the vision of the organization. I think IDP really truly represents this vision of connect and protect. It starts with products that make our homes smarter. It's also products that are making our cars smarter. It's products that are connecting the global wireless infrastructure. It's also driving that insatiable demand for data. The instabilities around the world are also driving our defense business to continue to grow. Our products also connect and protect in ways you may not have thought about. We supply products to the satellite industry, to companies like Hughes and Viasat and Harris. These companies are distributing data around the world that really help us connect no matter where we are. We also supply products to the automotive industry that makes it safer for us in our day-to-day driving, including products that are enabling things like autonomous driving vehicles.
Bob talked about a couple of years ago that we optimized the portfolio in IDP. We did this with a focus on balance of both growth and profitability, and we continue to have a very good run inside the business. We've experienced several years of greater than 20% growth, and over that same time, we've doubled our profit. On a go-forward basis, our core markets, Wi-Fi, the connected home, the connected car, the base station market, optical, and defense, we see growing at an estimated compound rate of 10%-15%. These markets also have some very positive underlying growth trends. GaN adoption is continuing at a feverish pace. I'll talk more about that going forward. It's deploying inside the defense, broadband, or cable, and base station markets. 5G is coming perhaps even quicker than we thought. We'll talk about trials around the world.
I'll show you products and what we're doing to enable those trials. IoT is just a vast number of opportunities for us to grow. Today, we focus mostly on opportunities in the smart home and in the car. To be honest, there's more and more opportunities that come up every day that our products are able to support. Let me talk a little bit about these underlying markets and how we model the market. We do believe that we've got a very nice foundation for growth as we go through the next five years or so. You can see here that, if you go through this, it represents that double-digit growth that I talked about earlier. The chart shows significant growth in the base station area. That's driven by the adoption of both GaN and 5G.
GaN will come a little early, 5G will start to deploy later. I'll talk again about those in slides coming up. IoT is really driven by the connected home and the connected car. Both of these markets will drive substantially higher than the average rate that I talked about earlier. We see in optical a return to what we would call a normal business. As most of you know, the optical business has really struggled over the last couple of years. One of the great things is we've been able to grow 20-some-odd% with that headwind. We do see optical starting to return. Broadband has been a relatively small market, been growing at a very steady pace. With the adoption of GaN, in parallel with many of our production programs starting to ramp in defense, we just continue to have very strong defense business.
I said earlier I wanted to talk about the strategy of the business. For me, there's really three things that you test your strategy against. One, does it create a competitive advantage in the market? The second, will it stand the test of time? The third, probably the only one that matters is, does it actually produce the results that you desire? I think we've been able to hit all three of these. I believe our technology, our partnerships with our customers, and the products that we've been able to produce have definitely created a competitive advantage. They've also allowed us to move into new markets and create competitive advantage as well. As far as the test of time, the strategies remain relatively unchanged over the last several years. With our view of the markets today, we do not expect that to change much.
Does it work? Clearly, as we reposition the portfolio and adopted this process of looking at the strategy, it's definitely worked. We've had a phenomenal couple of years. The strategy is relatively unchanged, I think it's working very well. What I'm going to do over the next couple of charts is I want to walk you through a little bit of details on each one of these topics. When we talk about our partnership with our customers, first of all, very impressive list of customers. Some of that is because we cross so many different markets. A lot of that is because we have so much technology that these customers need. We've got a great range of customers as well. It can range anywhere from the very large customers that you see on the chart, like Huawei and Northrop Grumman.
It can move into smaller customers like Garmin. We can move to very established customers like Ericsson and Cisco, also to new, very quickly growing customers like eero and Phicomm. Why do they come to us? Because we've been able to earn their trust, we've been particularly able to solve their most challenging RF issues. It could be putting GaN into an advanced defense system at Elta in Israel. Very challenging problem, we've done that very early on with them. It could be implementing coexistence filters using BAW into distributed Wi-Fi systems. It could be partnering with UEI to develop Samsung's next generation of RF remotes. Definitely, our customers count on us to solve their most challenging issues. Let me talk a little bit about technology.
We built IDP, in fact, Qorvo, really on a foundation of the industry's leading semiconductor technologies. Most of these are internal, some of these are outsourced to strategic partners around the world. You see the list there. GaN has been a staple in many of our businesses. Continues to drive significant growth. We've got advanced GaAs processes as well that cover everything up to millimeter wave. We're able to use BAW and SAW in many applications, we continue to push BAW to higher power levels and higher frequency levels to meet the applications for IDP. We have access to some of the best silicon technologies in the industry that let us do things like reduce power and cost in some of the product sets that we have.
We take those world-class semiconductors. We layer on some of the industry's best packaging and assembly and test capability. We have a very broad range of packaging capability, whether that be overmold, it can be air cavity if it's higher in frequency. It can be LCOR, which is a low-cost organic packaging for us. This has particularly been developed for some of our GaN applications. We take that packaging capability, and we put high-speed test and assembly operations around it. That's very important because some of the IDP customers demand a significant amount of test. We have to have capability to do broad testing. As an example, if you look at our optical products, we have a broad variety of tests that have to be created around each product that's delivered. In many cases, we've now created a product.
We now integrate that. We use multiple technologies integrated into complex modules. In a lot of cases, we take firmware and application software and layer that on to develop a full solution for our customers. This is particularly true in our low-power systems business. Cees will talk more about that coming up. All focused around meeting the most difficult challenges of our customers. Let me talk a bit about diversification. I talked earlier about being balanced and growth and profitability. This diversification has really brought us stability and predictability. We've been able to survive the ups and downs of multiple different markets as we went over the last couple of years and deliver consistent results quarter after quarter, really due to our diversification. We're diversified in many ways.
We have over 5,000 active customers, and they're all important, but none of them are over 10% of the revenue of IDP. We also have 6,000 active products that cover all of the technologies I talked about earlier. We service six different markets. Those are balanced in that growth and profitability that I talked about earlier. I'll talk more about us continuing to add products. We also continue to assess whether there are markets where our technology will play in a broader role. Let me now go in and talk about products. I want to talk about the concept of market-shaping products. These are products that we think create a competitive advantage for Qorvo and really move a market towards the kind of technologies that we have to play.
We call this our product development engine. It really has driven our growth. Last year, we released 122 new products, many that we think changed the game. We were the first in the industry to release 39 gigahertz FEMs. You can see the quote there that addressing the challenges of the next generation of millimeter wave. Very big challenges in this industry to move up to very high frequencies. Interconnects are challenging. Everything gets much, much smaller. We need to be able to integrate our technologies in much smaller package format to make these systems possible. We were also the first in the industry to release a multi-protocol, low-power system on a chip and recognized there as the most innovative product in 2017. We continued innovation with the first PA for vehicle-to-everything automotive connectivity, and we increased our BAW handling capability by 4x.
In IDP, the reason BAW handling power capability is important is it allows us to service these massive MIMO systems that you see starting to proliferate in the base station market. We then moved on to distributed Wi-Fi, and Cees will talk a lot about distributed Wi-Fi. We believe this is a trend that's really taken hold inside the industry. In a large part, we've been able to enable that industry with the use of our BAW filters integrated into our films. This really allows for maximum use of the Wi-Fi bandwidth, allows them to segment the bandwidth, and move data more efficiently around your home. Then in GaN, we released a 1.8-kilowatt transistor. It wasn't too many years ago that getting, of course, in my career, getting to a watt was a big deal when I started.
It wasn't too many years ago that getting to 100 watts was a major achievement, today we're doing 1.8 kilowatts in one chip. Amazing advancements. I think you can see, if you think about these products that have been released through the year, that many of these will really change the way markets are implemented, and we think that gives us a significant competitive advantage. I talked earlier about the trends, we want to spend some time, Cees and I, talking about these underlying trends and how we think they're going to drive growth inside of IDP. I'll talk some about GaN, I'll talk some about 5G, then Cees will get up and give you a broad view of what IoT looks like. Over the last several years, I've been very passionate in talking about the adoption of GaN.
I've been projecting over the last several years that GaN would grow somewhere in the 20%-25% range. Turns out I was probably wrong. It looks like it's actually going to grow faster than that. The projections that you see here on the chart show that growth rate is actually a little bit over 30%. It's driven by numerous things inside the defense and base station market, very exciting times for those that have this technology. If you look at why GaN, we talked about this before, but what's really driving the adoption? The short answer is it's creating significant customer benefits and value for our customers. We see very similar value across all the markets that we serve. Those values are things like lower operating costs.
The efficiency of the device has really helped our customers put systems in place that use significantly less power. I'll show you an example coming up of an antenna array or a massive MIMO antenna that uses 40% less power than an array implemented with silicon germanium. There's other examples, like one of our defense customers that was able to save $3 million a year just in fuel cost to be able to operate an array. The difference in efficiency between legacy LDMOS and moving to GaN saved them $3 million a year. Capital cost is also very, very important for our customers. The deployment of new capabilities, whether that be in wireless infrastructure, in defense, they need to drive their capital costs down. You'll see an example coming up again of where array sizes are 20% the size, one-fifth of the size of a SiGe array.
You can imagine significantly big cost reduction. Another driver is reliability, 100 times the lifetime of conventional technologies, whether that be GaAs or LDMOS, 100 times. You think about maintenance and repair and capital replacement cycles, significantly improved as we go through with GaN. What's driving most of everything is the ability to drive more data through our systems. We can do that with GaN in many, many ways. We can go to antenna concepts like massive MIMO and make those real, and I'll talk about those, but we can also go to broader bandwidths and higher frequencies all the way up to 100 gigahertz. Why are we going to win in GaN? First, we've got a tremendous legacy of innovation, partnered with the DoD for many, many years since 1999, 1998.
We continue to win significant amounts of funding from the Department of Defense on increasing our capabilities. We've got a broad set of solutions, anywhere from very low frequencies to very high frequencies, and in fact, we release about 60 products a year into GaN. We've been able to scale the technology, 9 million products, not transistors, 9 million products delivered, and we continue to scale to meet the increase in demand that we see from our customers. We've got tremendous market presence. We're a top three in all of the markets that we serve with GaN. In fact, we're number 1 in several of them, and we're a trusted foundry by the DoD. This market presence, combined with all the other components that we supply in these markets also, we think, gives us a significant advantage. Let me shift to 5G.
I've talked a little bit about the base station market when I talked about GaN, and it will definitely be a significant growth driver for base stations. The next one that we'll layer on, or the next wave is going to be 5G. We see the below 6 gigahertz, you see it there in black, below 6 gigahertz massive MIMO solutions starting to come on board. We're in development now, and in fact, we began delivering some production orders, and we really expect those systems to start to deploy next year, and certainly continue to ramp as we go forward. In millimeter wave, we're also in development now. In fact, we participate in numerous trials around the world. We do expect millimeter wave to ramp up also probably about a year or so later. What do our customers need to implement 5G?
I think you guys know why 5G is important, what do our customers need to be able to implement it? Well, first, they need to move to new frequency bands. Most of this work is going on between 3 and 6 gigahertz. Those are new bands. They need new products. They need new capabilities. It's one of the reasons you'll see coming up that we're developing BAW that moves up into these frequencies. Of course, millimeter wave, 28, 39 gigahertz, and we've got a broad set of products that support all those new frequency bands. They definitely need smaller form factors and lower cost. Predominantly true when you get into millimeter wave. What we've been able to do is those words there about technology optimize.
That means because we have all of this technology available, we don't just have to produce an array that's SiGe. We can use GaAs, GaN, SiGe, all these different technologies to really optimize the solution. We developed integrated multi-channel modules that allow our customers to package these in a much lower cost, smaller footprint. Of course, GaN will drive significantly lower power consumptions, I talked about earlier. I'm going to show you an example of that in the next chart. This is an example of two active antennas, two massive MIMO antennas, however you want to talk about them, both delivering radiated power, EIRP, of about 65 dBm. We think that's the requirements that will happen in the future. Both these antennas have very similar capabilities. The one on the left is implemented with SiGe only. It's got 1,000 elements.
That's important because that determines the size and the cost. You can see the square millimeters, about 4,000 square millimeters. We took those same requirements and designed a GaN-based array. It's not all GaN. It's got GaN power amplifiers on the output, it's got SiGe beamformers that are back in the chain. We've been able to reduce that count down to a little bit under 200, so about a fifth of the element count. It's a significant reduction in size. The end result, 40% less power consumption, 94% smaller die area. For those that think it's cheaper to do it in silicon, 94%. It has to be 94% cheaper in order for the math to work. We believe 80% lower implementation costs.
This is the kind of things that we think, again, will change the market or shape the market towards our technologies. Here's some great examples of those kind of products as well. These are the 5G FEMs that we talked about earlier, released in both 28 and 39 gigahertz. The smaller packages are single channel, the larger packages are dual channel. These are full up FEMs, so transmit, receive, and the switching in between. Today, some of these are actually in production deliveries. Many others are recent released and in the beginning sampling process, but significant interest from our customers. Now let's talk about how we're going to win in 5G. Very similar set of circumstances to GaN. We believe we've got innovation here that's going to allow us to win.
We've got all the technologies necessary, whether that be the semiconductor or whether that be in the packaging capability. It's no small feat to package at 39 gigahertz, so I want to make sure I emphasize the challenges at moving from two or three gigahertz to 39 gigahertz. We've been in the millimeter wave business for many, many years. It's quite the hurdle. We've got a broad section of products, many of these integrated to help our customers integrate their products more easily. Significant scale, with 150 million base station products shipped per year. We're very early in with things like the Olympics. We've been able to demonstrate this technology very early in the process. Strong customer relationships, we've got a leading share in base station small signal.
Those two combined, we think gives us very good position in the market, and we're seeing that by winning above 3 gigahertz today, whether that be 3 to 6 or millimeter wave. Next, let me introduce Cees Links. Cees came to us a few years ago when we acquired GreenPeak Technologies. I'm delighted to have him as part of the business. He's lived through the implementation of Wi-Fi in the 90s. In fact, was a pioneer in the industry, really solidifying the first opportunity for Wi-Fi to go commercial as it got put into the Mac in 1999. He's lived through a very challenging market with Wi-Fi in that time and consolidating down to standards. And the great thing is we have him on board to do it again. We're going to do the same thing as we go into IoT.
When Cees talks to you about this journey, it's not something that he's made up. It's actually something he's lived over the last couple of years. We're going to show a short video, and then Cees is going to talk to you more about IoT.
Ever been sitting on your couch and wished you could microwave a bagel, lower the thermostat, and fly a drone without getting off your butt? That, my friends, is the future of the Internet of Things, otherwise known as your phone is talking to your toaster, is talking to Pongo's food dish. Before your pup can have his toast while you get yours, IoT has a little growing to do. Let's get nerdy for a sec. The future of IoT relies on three main pillars: consumption, connection, and control. When we say consumption, we don't mean all those burgers you had on Saturday night. We're talking power draw. For IoT to work without being super annoying, sensors need to last a long time. Imagine if you had to replace 37 smoke alarm batteries each year. Exactly. You need solutions that last a decade of battery life.
The pyramids of sensors, if you will. Connection right now is a little complicated. It's more of an all-steel cage match between Wi-Fi, Zigbee, Bluetooth Low Energy, Thread, LTE Cat-M, NB-IoT, and about 1,000 other protocols, all vying to become the standard. With all these protocols and no clear winner, the only way to be safe is to take a multi-stack agnostic approach, which just means technologies that work with any protocol. Just talking to everything isn't enough. You need control, or more importantly, intelligent control. On their own, these devices are just things. When linked through control software on your phone or in the cloud, all these devices work together like a genie or tiny kickass virtual butler, securing your home, improving your health, and triggering a paintball cannon every time your neighbor Gary lets his dog unload in your yard.
Come on, what are you feeding that thing? Now, enjoying sweet retribution from your office, that's control software doing its thing, connecting you to cannon from miles away. IoT technologies are destined to move into everything that we use, from diapers to cities to regional power grids. We're talking smart phones, smart cars, smart keys, smart doors, smart cameras, smart trees. So much smart you'll be tired of smart things. We're not joking. That word will get used more than millennial, trust us. The Internet of Things connects people and their devices to one another in new and powerful ways that simplify our lives. That means more time for family, friends, and that Kickstarter project in your garage that your boss totally doesn't know about. Excited about IoT yet? You're welcome. Want to learn more? Talk to Qorvo. Now, excuse me while I feed Pongo a treat.
Good boy, Pongo. Who's a good boy?
Good morning. This is probably the 23rd IoT presentation you will hear today, bear with me. What I really want to share with you is not so much about IoT, but a little bit more about what we are doing in IoT and how we think we can make a difference. James said already earlier, what really makes a difference in companies, and in particular in IoT companies, is the strategy and how you execute. We are living in a very interesting time. I will show a little bit more about cars in a minute, but just remember the time that a car was on four wheels to bring you from place A to place B. A car today is not something to bring you from place A to place B. It is about driver's license. It's about rules. It's about law and law enforcement.
It's about insurance and liabilities. It's a whole world. We are still in the internet, in the Internet of Things, in the very early stages of what we want to do with the internet. I would like to lead you a little bit through our thinking, which starts to pull many of the things that we are dealing with today together and actually makes it very exciting. I would like to run you somewhat through the rapidly growing markets, but also how we are addressing the markets, a rapidly growing market can be very chaotic. What we are seeing today is very chaotic. I can imagine you can get confused very quickly about what is all playing here. In particular, when we talk about all the multiple standards.
The multiple standards that are playing in the Internet today probably take care more of confusion than they help to grow the market. Our whole philosophy about the market is not so much like let's pull everything together and do everything for everybody. It's actually way more, what are the winning applications today and tomorrow? Because when we really think about Internet of Things, it's about creating value by being able to make better decisions faster. Providing data and providing data at the right moment and at the right place. We in Qorvo see ourselves very much as facilitators. How can we bring data around in an efficient way at any moment in time that we need it? When you think about that, you realize that connectivity, but also system-level thinking is very essential in the way we deal with the Internet of Things.
You've seen probably many graphs with growth and growth charts. Probably the best takeaway when we look at Internet of Things, we look at a piece of the market that is very much focused on low power. You have heard already in many of the statements earlier, low power, long battery life, is very essential for what we are doing. I'll run you through a few of the key items that we really concentrate on in development. You probably can relate to your own house. You can relate to your own activities in your own house to see how these things connect. Like you saw in the video, changing batteries is probably getting as ugly as changing light bulbs.
Fortunately, we have now long-living light bulbs, now we get all these devices with batteries that if you don't have to change the battery, then at least you have to recharge the battery, which is equally a nuisance. From the IoT perspective, we look in the first place at long battery life. Actually, we have split the market up in three segments. The first segment is the short range, the second segment is local access, and the third segment is wide area access. Actually, when you think about your phone, it's not that strange because in your phone, you have three radios. It's a Bluetooth radio for local connectivity. It's Wi-Fi for being on the network at home or in a hotspot, or if you're wandering outdoor, it's the connectivity that you have connected everywhere in the field.
These three radios we think are also the precursor to how the IoT will develop in just three simple radios. One radio focused on very low-power connectivity, one radio focused on very low-power networking, and one radio focused on, say, outdoor cellular connectivity. The only difference with what you're used to in your phone, in your phone, it's really about high data rates. We talk about IoT and IoT connectivity, it's in first place about long battery life. Actually, it's fairly straightforward, but that's only the beginning. I want to run you a little bit through the history of where we are coming from and where we are going. You see this chart, the horizontal axis, you see from 2000 to 2030. We keep a broad view. From the vertical axis, you see data rates. It's amazing.
When we had the first Wi-Fi 20 years ago, the data rate was 10 megabits per second. Let me tell you something. When we did our first Wi-Fi, we were competing with 3G. I was working at that time for a company called Lucent Technologies. Lucent Technologies was all about 3G, and 3G was four megabits per second. You know what my boss to me said? He said, "Four megabits per second, that's enough. People won't need any more data rate for a long time." You can a little bit see what happened on this chart, and data rates are going up. They have been going up for years and years. Our expectation is that the need for data and data will go up.
What we are doing in connectivity and in Wi-Fi is very much leveraging data rates going up and the new upcoming standards, Wi-Fi 802.11ax, sort of the Wi-Fi equivalent of 5G. That's something that we are spending a lot of time on. The reason I'm talking about Wi-Fi 802.11ax because we see very nice crossovers between Wi-Fi 802.11ax and low-power connectivity. If you see the bottom line here, you see the bottom line is the low power line. Actually, in data communications, you get one or the other. If you get high data rates and you keep the range the same, then the battery life goes up. What you see on the lower line is the line with the long battery life, and on the higher line, that's the line with the high data rate.
We all are familiar with the high data rate, I'm talking a little bit more about the low power line, which initially started with Bluetooth. You've seen Zigbee emerging, you've seen Thread emerging, and you see the applications that are kind of emerging with it. Still by the day of today, there is no end to the need of applications that want to use these data rates. Interesting to mention, there is a kind of a break, there's another line coming up a little bit higher. It's a standard that exists already for a while. It's in the millimeter wave. It was called WiGig. It's now adopted in the Wi-Fi Alliance. If you look at where the data rates are going, that could be coming a very interesting technology in the coming years as well that we seriously want to keep an eye on.
Put it in perspective and where we are today, it's pretty ugly. I've done my best to try to find as many standards as I could find, I hope that I don't insult anybody by forgetting one. You see, again, short range, local area, wide area, you see content, sensor control, proprietary solutions, then all the standards that are playing in that space. Don't worry if you don't comprehend this whole slide in one go. It's not that important. I'm just trying to indicate how much chaos there is right now that there is a lot of competition going on in terms of companies that are really pushing their standards to become the dominant standard in the industry.
In Qorvo, we have a view on this because ultimately we are an RF company, and ultimately we want to serve all our customers in providing the best RF we can provide. In the way we see this market evolving, it's actually probably becoming as simple as your phone. With the three radios, Bluetooth, Wi-Fi, and LTE, we think that the low-power world will probably split up in Bluetooth, in Zigbee, and in Narrowband IoT. The simple reason is that it is very expensive to keep all these technologies in the air, and open standards in communication over the years have shown to be kind of consolidating a few and a limited number of standards. You see that I have there Zigbee and Thread together in one box. Interestingly enough, they both have the same radio.
Actually, they are brother and sister, and I have my views on how that will develop. Frankly speaking, I think you probably don't even have to make a choice. The industry will make a choice for us in the way we have to deal with this. Here you see again the picture. Bluetooth and Bluetooth Low Energy, Wi-Fi and Zigbee, and LTE and Narrowband IoT or its niches like Cat M or that type of technologies will play in the space in cellular area. This is our view in where the market is going, and I would like to play this out a little bit in terms of what does it mean for Qorvo. In Qorvo, we are very selective in where we want to play. We realize that being everything to everybody is not necessarily the best return for our shareholders.
We try to be really focused on where we can win and how we can differentiate. Interestingly, Bluetooth and Zigbee, the first and the third column, we are very much focused on ultra-low power and very strong connectivity all through your house, all through your area, your local area, and have a very long battery life in that space as well. Actually, we are focusing on battery life in that space that exceeds the life of the product. Because ultimately, we believe that making something maintenance-free is the key for all the products in the IoT space. Because just think about it. If you have 100 devices in your house in a few years from now, if the battery life is a year, you will change two batteries every week, and that's no fun to play with.
What we also see in this space is Wi-Fi and LTE and Narrowband IoT. When we look at these two technologies, that is very much the core space where we started in Qorvo, with frontends, where we started with all the portfolio products that you have seen and heard of before, how we differentiate on filtering and how our BAW filters are really a key opportunity that we are using. Because what's happening in this space is that we will see more and more radios getting together and getting together in the same area. The last thing you want is that one radio interferes with another radio, because that is really causing trouble. I already promised you I was going to make a little step out into the automotive market. I want you to close your eyes for a second and go back 25 years.
If you go back 25 years we get a call. If you go back 25 years you get the choice, where would you like your automatic door locking? Would you like it in your car, where the doors are very close together, or would you like it in your house? If you had a choice, where would you prefer? Well, interestingly, most of the people would prefer it in their house, right? A universal experience is standing at the front door and not knowing whether the back door is locked. Where are we today? This is 25 years ago. We have 25 years central door locking our cars. You cannot even buy a car anymore without central door locking. Look at your house. Anybody here has central door locking in his house? Her house? Don't you think we are living in a very primitive world today?
That's just amazing. It's very interesting. We believe that automotive is a sort of precursor on what is going to come and what is going to develop the coming years. We believe it's a very exciting time. If there is anything primitive today, I have to say, the internet that we are using today is very, very primitive. Let me run you through what we think that the internet is going through. Actually, this is probably, for many of us, the internet situation at home. You probably have one router somewhere, maybe at the front door, maybe behind the TV, or maybe in the study. You have your family members complaining about range and connectivity because you don't have good connectivity in the backyard or up in the attic or down in the basement.
You're struggling with extenders and all kind of equipment to get this kind of problem resolved. Well, the interesting part is the new standard 11ac and 11ax are really resolving that problem already. That is called distributed Wi-Fi. You don't go to the store anymore to buy one router. You buy one router and a set of boxes that you can distribute through your house and that wirelessly connect with the main router at the front door. This is called distributed Wi-Fi. Think about it from a Qorvo perspective. It's a kind of a dream, right? Instead of one router, we have now four boxes that wirelessly connect. At the same time, when we go from 11n to 11ac, we go from one by one MIMO to four by four MIMO.
The number of power amplifiers and the number of components that we are selling to router vendors is multiplying, not only because the technology is multiplying, but also the number of boxes that are placed in the house are multiplying. This all has to do with what people really want in the home. They want to have good coverage. They want to have coverage in the attic. They want to have coverage in the basement. They want to have coverage in the backyard. With higher data rates, it's really a struggle to get that coverage. From that perspective, we really see distributed Wi-Fi as the horse to bet on for the coming years in terms of really focusing on winning the design slot and growing our business.
The real thing is to come still, because what's happening on top of that is a sort of wireless RF explosion. Today, probably, and this is an interesting statistic, you probably have in your house maybe 10 Wi-Fi nodes. Think about it. Some PCs, some tablets, some phones. Say 10 Wi-Fi nodes. I remember the days that we were trying to sell the first Wi-Fi into the first computer in the home. I can tell you, it was a struggle to get the first Wi-Fi sold. Nowadays, you probably have 10 Wi-Fi nodes in your home. We think that the coming years, the number of devices will probably grow to 100 or beyond that. All these devices will require RF in one way, shape, or form.
Even if you give it a closer look, actually, you will see the home become a sort of telephone with all the different radios that are involved there. That we need to take care of that interference is not disturbing all the signals that are playing a role there. That is where we see the interesting thing going. There's something else if you really study this picture. You see a little microphone. Remember Alexa? Kind of a separate microphone in your house that you can ask what the weather is outside. My wife always says, "Why don't you look out of the window?" Nevertheless, I like to ask Alexa what the weather is outside, and she nicely tells me. We think that Alexa will be integrated in every pod in every room. You can talk to the internet in every room.
We think that there is a complete wireless explosion, including voice. Look now at every pod. You almost recognize in every pod a cell phone because it has Wi-Fi, it has Bluetooth, it has Zigbee, and it has audio capability. That remote control from Comcast that you're talking in today to order a movie, it's just the start. It's a start of a household that will be completely changing, and that is very exciting. You all will be witnesses of that while that is unrolling in front of your face. What are we doing in this space in Qorvo? I already warned you earlier, we are not trying to fill all technology for all the people. With this vision, we see where we really want to focus, where we want to win, and what gaps we have.
With other words, with whom we want to partner for filling these gaps and working on the next generation. You see our front-end business, you see our module business, where we are very strong, both in modules for Zigbee and Bluetooth as well as for Narrowband IoT. You see our system on chips, in particular in the low-power space, because there we have already a lot of integration. You see the firmware and the application software, and you see the systems. What's very interesting, nowadays, and that's really a change in my business unit. We are selling systems. We are not selling components anymore. We are really selling systems. How do all these radios work together in a cohesive way and in a way that things really work? What is this all about? Let me quickly summarize what we are talking about.
We're talking in Qorvo about superior capacity. Being able to support many devices using many different radios at the same time. At the same time, we talk a lot about a high level of integration. It's very funny because the key word that we are using for this is small is beautiful. If you have a router box in every room, my wife would never like such a big box in every room. Can you make it smaller? The real thing is if you make it smaller, you're seeing all these PAs that radiate a lot of heat. How did it get rid of all the heat if it's so small? The best way to get rid of heat is not to generate it.
That is where a lot of our technology, whether we talk about PAs or whether we are talking about filtering, is playing a role. Energy efficiency is not only important for long battery life, but it is also very important for small products. Believe me, if you go to the store and buy a router, you have the choice between a big router or a small router. I think what you really will start liking is these small routers that really help you connectivity throughout the house. The last thing, of course, and that is really something that is a really strong point in Qorvo with our filtering and our BAW capabilities, that is how do we separate all these radios from each other? How do we avoid that our radio signal gets interfered by the radio signal of the neighbors?
It gets interfered by all kind of other stuff that is out there. Filtering is very key in this whole space, and we are really playing a major role there. Summarizing this all for IDP, we are really working very hard to build a new future where we enable people to make better decisions faster. You being here, it's really fantastic, so you can be with us and be a witness of what kind of future is unfolding itself, and all the great things that we are trying to solve. This is the last thing I want to say. Thank you very much. We have a break now. We expect you back in 50 minutes, 9:40 A.M. We expect you back in 9:40 A.M., yes. Thank you very much for your time.
Right here. Todd.
We're about to get started pretty soon. If everybody can please take their seats, that'd be great. If everybody could finish pouring your coffee out in the hallway, we're going to begin in about 30 seconds. All right. It's my pleasure to introduce Eric Creviston, who is the President of our Mobile Products group.
Good morning, everyone. As Bob mentioned in his opening remarks, it's very rewarding to work in an industry that's impacting so many people's lives. There's no question that the effect of mobile data connectivity has been profound, and it's showing no signs of slowing. As more and more people get connected across the world, as applications are ever increasingly using more and more data, and in fact, video is driving an awful lot of the data consumption, as I'll show you later, it's requiring an exponential increase in mobile data delivery. As we've talked about many times, and we're going to talk about over and over in my presentation, that fundamentally requires more and better RF, and that's exactly what Qorvo's focused on. At Qorvo, we've spent the last few years building not only a broad technology portfolio, but a very deep technology portfolio.
We now have all the pieces we need to address the fastest-growing and most impactful parts of our market. We're very happy today to be sharing with you our plans for how we're going to attack this market. Let's start by talking about the market itself. Today, we're serving about a $14 billion TAM for RF components. We see that growing to over $20 billion within the next four to five years, that's due to two primary factors. Clearly, as you all understand, each successive generation requires more RF, the content goes up in each generation. Also, as users upgrade their devices, you're seeing a mix in the overall handsets towards handsets with more RF in each one of these years. That mix of users affects the total RF TAM substantially.
That's why the RF TAM can grow in dollars significantly, even as handset units are flat or even down, because of the mix shift and the increase in content generation over generation. We're showing you here how we're modeling the current legacy platforms of 2G, 3G, and 4G declining and being replaced with two waves of growth that we see coming. The first, which is really upon us right now and what we're working very hard with our customers on, is LTE Advanced and LTE Advanced Pro. LTE Advanced Pro getting you to the ultimate performance in LTE. It's the gigabit LTE phones which people are announcing, and we've been using our parts to demonstrate gigabit LTE for over a year with our leading customers.
That brings 4x4 MIMO, 256-QAM, a lot of carrier aggregation, all those things we've talked about, which expand today's RF content in LTE-Advanced and Pro. We've got line of sight to 5G. 5G is probably the most potential upside that we've seen in the industry because it's the first cellular standard that addresses more than just high bandwidth data connectivity. As you know, 5G cellular will bring two additional modes, one low latency, high quality standard, which will affect autonomous driving, augmented reality, and so forth. Then also, as Cees has already covered, a mode of communication which is very long battery life for IoT sort of applications. This is very different than previous cellular generations in that it addresses multiple market segments. There's a wide range of possibilities of how high 5G could go. It all starts with mobile data.
I've said this every year we've done these things, I always start by saying it starts with mobile data. That is fundamentally what you have to watch for. Look at the Ericsson reports, look at the projections for the demand for mobile data. You'll see every year it continues to get higher. It's an exponential curve, we don't see that changing. It's being driven more and more, not only by applications where people are sharing video, but also live streaming direct video is driving it. Video, in fact, is currently growing at 50% a year. The amount of data consumed by video over the mobile internet is growing at 50% a year, and it's projected to be 75% of the market within a few years. That's because, as I covered last year up here, the video communication is far richer than static photos or obviously text, right?
If you look at the real-world example from last weekend's wedding, you see all those outstretched arms there. They're not just waving hi to Prince Harry. They've got phones. If you zoom in on this picture, they're holding phones up. Some are taking pictures, but I'll bet you a lot of them are live streaming video real-time. That is the way people are communicating these days. If you can only imagine how much data it takes to service. If you pan out on this and see how many people were in that crowd, it is really a phenomenon. It's happening all over the world at sports events and so on and so forth. Fundamentally, this demand for mobile data requires more and better RF. To go one layer deeper on why that is, it starts with all these market requirements.
If you want to get more data through the cellular infrastructure, you have to implement a lot of higher complexity RF functions. 4x4 MIMO is where you've got four independent transmit and receives on the phone, four separate antennas pulling down four parallel data links. Each of those data links have 256-QAM modulation, much higher order of modulation than on most of today's handsets, allowing more data to go into each of those channels. In addition, more and more carrier aggregation, where each of those channels is carrying multiple carriers worth of data. You put all those together, that's how you get to the gigabit LTE we talked about. Now there's more bands being added as well, going down to 600 MHz and going up to over four GHz in ultra-high bands.
New frequencies, new modulations, much more complex front-end architectures, all of that is how we're delivering the mobile data that's required. The problem statement is the form factors are getting more challenging. You might think the phones are getting bigger, so there should be more area available. First of all, not how that works. The RF never gets more area. They got too many other things to spend their space on. It's even worse than that with this industrial design trend of taking the screens out to almost the entire edge of the package. If you look at this example, the phone on the left is smaller, but it's actually got more area for the RF, especially the antennas, because anywhere where there's not screen is where you get to radiate RF power.
There's a lot more black, if you will, in the smaller phone. The larger phone with the edge-to-edge screen, you see there's almost no place to put those antennas, despite the fact that the trend I just talked about says we want more antennas. We want more parallel channels running. We want more frequencies. We need more antennas to cover more range. All of that in a much, much smaller area. That is a massive complicating factor for our customers to deal with while trying to pack in all this RF capability. Lastly, power levels are going up.
Not only power class 2, where the phone is emitting twice as much power for Sprint and China Mobile in some cases, but also within the handsets, because as we deal with all this antenna complexity and we pack all these things in, you have additional losses between the power amplifier and the actual antennas. There's more power being generated inside the RF to overcome those losses. As Bob hinted at earlier, that makes our SMR BAW in particular very well-positioned to help deal with that, in addition to, of course, our state-of-the-art power amplifiers. It's becoming multidimensional. All of these complexity factors and challenges are building upon each other to make the overall thing exponentially harder to do, which, of course, is good news for us.
If we look at that TAM growth and peel it back a bit and see how this complexity affects TAM and take it to the product level, I know this is a bit of a confusing chart. What we're trying to do is take the roughly $14 billion of today's TAM and put that in the black bar at the bottom, then build up how we see this additional TAM growing. If you look at LTE Advanced Pro, adding 4x4 MIMO and 256 QAM and so forth, as I mentioned, that's really affecting the PADs, which are power amplifier duplexer blocks. These are the workhorse of the main path of the radio, the transmit and receive. Most of that's being integrated into highly functional PADs, as we call them. There's certainly value being added there.
There's also additional multiplexing added in terms of discrete components to handle some of those modes. As you go to 4x4 MIMO and more diversity, you get more diversity modules, of course, more content there. Then around the antenna, to handle all that complexity I talked about, you'll have antenna plexers, which allow radios to share antennas. Very important when you've got all these radios trying to work in that very tiny space around the edge of the phone. You have to share antennas with antenna plexers. Because they're very sensitive to the touch, you use a lot of antenna tuning to make sure the performance isn't compromised. Going forward to 5G, we see that adding about $1 billion in RF value beginning in 2020, about $1 billion per year in terms of 5G adder.
You see it's really a story of more of the same. Some of the multiplexing, in fact, most of what we believe by that time will be integrated into the pad section, you see that really outgrowing. DRX or diversity receive continues to grow, antenna complexity, especially with all the new bands 5G is bringing, will outgrow most others in terms of rate of growth. Since 5G is so important to driving our TAM in the next few years, it's important to understand how we see it rolling out or going to market. I think as James indicated, if anything, we're seeing it sooner. It's here today for IDP, for sure. Serious base station design is underway. Also for mobile. It's been over a year. It was at Mobile World Congress 2017 when we announced the world's first 5G RF front-end module for mobile.
By now, that module has been used in just dozens and dozens of field trials by all of our customers. It's getting a good workout in terms of proving what 5G can do. We're still in the trial period today. The specifications are released for NSA, which is non-standalone. In the U.S., that'll be the first mode that'll be rolled out. You'll have some trials this year and some models going to market in 2019. Could be handsets, could be dongles, could be more like pads. We'll see some volume in 2019 on non-standalone, where you still use the LTE backbone with just the 5G protocol running the radio on top. Then next year, we'll get the full specs for true new radio 5G, not standalone, but stand on its own 5G. Those specs will be released next year. You'll start trials of those 2019, 2020.
By late 2020, we expect to see mass volume of true 5G new radio. As you go from 2020 to 2022, of course, you'll see that begin to go and propagate down throughout the tiers to where by the time we get to 2022, we've got true mass production, very mature technology, and the majority of the phones beginning to run 5G protocols. When you add 5G to the phone, this isn't something you can just bolt on to the side. You can't take a 4G phone and add 5G by just adding a 5G radio to it. It really impacts a great deal of the 4G system as well. This block diagram is attempting to illustrate the fact that there are some new blocks, the ones that are solid blue, which are running true 5G only protocol.
They're 5G specific content, also a great deal of the 4G content, especially in the PAD area, is being affected because it now, even though it may only be processing LTE, it has to coexist with 5G and those other frequency bands, it has to run cleaner, more linear. As I mentioned, the added complexity, it might have to run more power level in those parts. 5G will affect a great deal of the radio, and you really need to co-develop all of these blocks together. This, we believe, is a unique capability of Qorvo, since we have every single bit of this block diagram in-house. How are we looking at this opportunity? Obviously, it's a great opportunity for Qorvo as well as for our entire industry. We believe that our primary role is to deliver the best RF possible.
We invest in a wide range of core-enabling technologies. You heard about that from Bob. You're going to hear a lot more from Todd and also Steve on that. Since we have the capability of doing everything, it's also becoming more and more important that we spend a lot of time on disciplined portfolio management to make sure we're focusing our resources on the areas where we can help the industry and help our customers the most, and of course, get rewarded the most for it. A key part of our strategy as well, just as it was with IDP, is partnering with industry leaders. This includes participating in the 3GPP standards. We help develop the standards for next-generation technology, as well as our supply chain partners developing new breakthrough technology, in both silicon processing as well as in packaging.
Of course, our customers as well and our go-to-market partners. I want to walk through these briefly. I won't say a lot about technology, because you're going to hear this in great detail from Todd next, and he certainly is better suited than I am to talk about it. At the highest level, as you know, Qorvo was formed to do this. When you look at the mobile technology portfolio, Qorvo was built to have all technologies in-house, and that is where we started. What have we done since then in the past few years? We've been working on depth of the portfolio, successive generations of each of these technologies that get better and better and more differentiated every year. We have the ability to put each of these parts together and build a superior total system. Here's the exciting part.
Once we have all of these in place, we've got proven capability that shows we're as good as anyone in any one of these product categories. We now have the job of sorting through them and targeting our investments for the areas that are the richest payback for us as well as for our customers to help them get to market. When we look at this, we stack the product portfolios, the opportunities on two axes. First of all, the growth rate on the X-axis. Although there is approximately 10% overall CAGR for RF TAM, there are some parts that are growing less than that and other parts that are growing dramatically more than that, obviously, right? We also look on the Y-axis at differentiation. Where can we be unique? Where can we solve the really hard problems?
What can we do by the fact that we've got all these technologies in-house and no one else can do this in a way that really helps our customers build better products, right? That's how we measure differentiation. When we look at this and look at the chart, you can see in blue, these are the areas that we're continuing to focus more and more on. In the lower left, in the discrete area, there are certainly some targeted opportunities. There are certain high-value filters, for example, that might still be sold discreetly. We'll still have opportunistic investments there, but far more of our attention is going towards the right of this chart, which is a highly integrated, highly differentiated part of our portfolio. You see right in the upper middle part there, the mid and high-band PAD.
This is an architecture which is being adopted across many tiers of our customers. Obviously, you've seen in the flagship smartphones that are out there, already moving heavily towards integrating mid and high-band functionality into one placement. That's because as you look at more carrier aggregation modes, they're more in the mid and high frequencies, and you put those all in one package, you can optimize better for much better performance and lower loss in the system. It's a natural thing to do. We see this carrying down into the performance tier as well. Our Phase 6 system, which is now released and in production, is for our performance tier. It's primarily our China handset customers that are looking to adopt Phase 6 architectures. Very similar split.
Mid and high band are integrated into one placement with all the filters integrated with PAs and switches, all tightly integrated for high performance. Then the low band integrated separately. We think the low band has got opportunities for us for sure. Our TC-SAW capability is very good. When we look at the market dynamics, though, there's about three-quarters of that market that's concentrated around one or two customers and one or two part numbers and three or four competitors. We like parts of the market, especially the Phase 6 part, where we can sell the low band combined with our mid/high to give customers a complete solution. That's something that's strategic for us. We can differentiate in that total solution, and that's something we should participate in.
The diversity receive module as well, the larger bubble just below mid/high band, this is an area where we don't currently participate, but it's a great opportunity for us. Especially as we look forward, it's going to become harder to do. More mid and high bands coming, less loss required, this will call for BAW filters. When we see the opportunity to bring our BAW technology in and complement our SAW technology in this footprint, that's going to be a great time for us to intersect this market. We're sampling today, expecting production late next calendar year. Antenna solutions, as you know, is a big part of our business today. We've been very strong there for many years. We're a pioneer in many ways around these solutions. We're going to continue to build on our deep portfolio.
Todd will share the differentiation we have in new generations of technology here. This is a massive focus for us to continue to help solve that antenna complexity problem I shared with you. Emerging ultra-high bands. These are the bands in the three to over four gigahertz range. A great opportunity for Qorvo, of course, to exercise our leadership in advanced technology. Putting it all together, we've got a broad and deep technology portfolio. We've got a clear portfolio management approach, which allows us to address the most valuable and most challenging parts of the market. We have all the right partnerships to bring these things to market, working with standards bodies, a very elite field application team, the best in the world, scattered close to our customers geographically as the key part of our strategy. We're putting this together.
I believe we're really hitting the ground running this year, hitting on all cylinders, and projecting a lot of growth during this year based on this strategy. To go deeper now into the technology, which is a key enabling part of it, I want to hand it over to Todd. He's our CTO and VP of engineering for Mobile Products.
Thank you, Eric. I got to make this thing go. There we go. I want to double-click one layer down. I can't go too deep. I think I was basically given 15 minutes, so we could spend 15 minutes on a partial topic up here, but we'll see where we can go. Qorvo has a very extensive, and I would say a premier portfolio of technologies. We either hold a number 1 or number 2 position in each one of these technologies. We don't just have a broad portfolio of technology, we've got very good technology. The BAW filters, we spent the last few years catching up, and we feel like we've closed the gap in the BAW technology, something I think we talked about last year. We made huge strides in our BAW technologies. SOI, we are the leaders in SOI technologies.
We develop our own technologies and drive our own technologies internally. This is something that we will maintain leadership on. Gallium arsenide, we're second to none. We've got very good HBT technology, we're extending some of our GaAs technologies to millimeter wave. Some help through James' group. We're leveraging some of the IDP technology to bring into these millimeter wave potentials for the handsets. Envelope tracking. We're one of the world leaders in envelope tracking. I believe there's two of us out there. Very important to manage the thermal dissipation in the phones. Very key. You don't want to be running very hot phones. Envelope tracking is a very key technology moving forward. I'll go into more detail in a little while. As Eric mentioned, all these new functionalities, these new standards, are all coming to the RF front end.
Packaging, being able to package all this new technology into these very small packages, very dense packaging, is very important for us to solve. The standards will keep evolving. 4G, LTE-Advanced, coming to 5G, new frequency bands, creating new challenges for the RF. Take, for example, 600 MHz, going down to low frequency. Low frequency for a handset is not necessarily a good thing for the handset. Brings a lot of value for you as a user, 600 MHz low frequency means big things. It means you want to have a big antenna, you want to have a big whip antenna, something like this. Remember these antennas we used to make? Yeah. You want to have big antennas, you want to have big filters, you have to get creative in how you solve that problem.
You bring in more antenna tuners, you bring in new SAW filter technologies in order to handle the 600 MHz. You move into the 4x4 MIMO carrier aggregation. That's getting rolled out. It is out today, it's getting rolled out across the world, that's going to bring more demand for multiplexers. You're going to see a lot more multiplexers, not just in the PADs. You'll see multiplexers move into the diversity modules. Modulation rates are increasing. 256-QAM is something you might see out there. Increasing modulation rates, that's to improve the data rates with the frequency bandwidth you have. What that drives is, for an RF guy, is what we call higher peak to average, higher output powers. Again, it drives the importance of envelope tracking to manage the thermal dissipation in the phone.
Every time the peak to average increases, envelope tracking can save much more power. The last standard, of course, that's the buzz. I don't know how many of you went to Barcelona, 5G. Especially the sub-6 GHz 5G, the new bands, the 3.5 GHz, the 4.5 GHz, is going to be driving some new technology developments, especially in the BAW acoustic area. We think there's a necessary need for BAW filters in these type of bandwidths. Filter, this is just a snapshot of the frequency spectrum that's available to a handset today. Starts about 600 MHz and goes up to 6 GHz of spectrum here. It starts in what we call the low band area, usually consists of using SAW and TC-SAW filters for these frequency bands of operation.
Around 1.5 GHz, it shifts up to what we feel is a good area for BAW. Typically, we've been using BAW filters from 1.5 up to 2.7 GHz, that's what's in most of the phones today. That's the split. Moving forward, we have these new 5G bands rolling out, what we call n78, n77, n79. These are 3.5 GHz bands, 4.5 GHz bands. As you notice, the bandwidth is very wide. Very wide bandwidth. Each one of those dots on this chart represents a band, a unique band. If you notice, the low bands are all bunched together, the mid bands are all bunched together. You go up in those 5G bands, they're very broad bandwidths and very high frequency. It sets it up to be very nice area for BAW filters to succeed.
The legend down here, we're sampling the 5G bands today, both the 3.5 gig, 4.5 gig, as well as down here, we've got one of them 600 megahertz bands. Very large breadth of filters needed for these handsets. Coexistence problems. I want to give you an example of why BAW is needed for 5G, sub-6, sub-6 gigahertz 5G. This is a benchmark of what Qorvo can do in its BAW technology versus what we call an alternative technology. BAW does have better insertion loss, more important than insertion loss, okay, I can live with the insertion loss for my handset, it allows you to use 5G band with your Wi-Fi. If you notice the chart on the right, with the other technology, you would not be able to use your Wi-Fi. It would desensitize your phone.
It allows you to coexist with five gigahertz Wi-Fi in this 5G band. This is a good example of coexistence problems we have in the handsets. BAW is the only technology can do this. This is the same block diagram that Eric had. I had to have one block diagram. They allowed me one. I wanted to take a little different twist on it. I wanted to focus on the two areas, the antenna plexing and the multiplexing, and why it's needed, because we think it's very important. As MIMO comes, diversity comes, 5G, this is driving the need for more antennas. The antennas could go up to 20 or 30 antennas, depending on how crazy they get. You could bring in antenna plexers. These antenna plexers allow you to reduce the number of antennas needed.
They provide, what I would say, a coarse selectivity per antenna. If you look at the n79, the MHB2, UHB1, that could possibly be three antennas, or we could just do what we call a coarse selector. It allows only those frequencies to pass through that antenna. These antennas are very broad. They don't care. They'll receive anything. They like to just receive. We need some type to coarse select, steer those signals down first to a coarse selector, then they go into what we call the PADs, and this is where the multiplexers come in, is they do the fine tuning. They'll come in, and they'll separate two mid-band singles to the receiver. There's a coarse selector, a fine selector to the receiver. Very important that these two coexist in the phone. Number of antennas are increasing.
Two to four today, probably going to seven to eight as these new capabilities roll out into the phone. Phones, from my perspective, aren't getting a lot bigger. I'm an RF engineer. Maybe I'm jaded a bit, Eric said they're getting bigger. Eh, a little bit. The number of antennas, the size of the antennas have to come down in order to fit in the phones. When the size of the antennas come down, they lose their bandwidth capabilities. They become much, much more sensitive. I put a cartoon on the right on the four, I think, use cases that are typically measured in a phone and kind of shows you what the pass band of the antenna does in the different use cases, how the pass band moves around, and it moves around more quickly as the more sensitive it gets.
What the job of the antenna tuner does is to recenter that band pass to make sure it's in the frequency of operation that you want. SOI. This is an area that we've been investing in for almost probably a decade now. It's an area that we feel is really key to maintaining leadership for antenna tuners as well as switches. Over the past five years, we took a snapshot, and we've doubled the, or more than 2x improvement in figure of merit. Ed will be more than happy to tell you what figure of merit is later today, I'm sure. I won't do it here. 2x figure of merit, but what it really does is allows you to have two-thirds of the same die area. It's a huge improvement in SOI technology over the past five years. Stepping on to millimeter wave.
I wrote this on purpose, 5G, millimeter wave, and mobile devices. Question mark. I think if you read the papers, it's coming. It'll be here, actually last year. If it comes, I think Qorvo's going to be ready. We've leveraged some of our technology from James' group, James' IDP group, and this gives you I'll walk through this in a second, but we feel like gallium arsenide has a significant advantage over silicon or SOI type technologies. What it allows us to do is it allows us higher output power per element, so we can reduce the size of the array by at least half. The demonstration up here, I'm showing from a 16-element silicon array down to an eight-element GaAs array. Or we could even reduce it further. We have interest. People are interested in maybe looking at maybe a two by two array.
What this is solving is size of the arrays. You can go into a handset. These handsets are very small. Sorry, I don't have my handset. Very small handset. We're looking at two or three of these arrays in each one of these handsets, areas are going to become very premium. Cutting the array sizes down by half or maybe even a fourth is going to be a significant advantage. That's something we feel gallium arsenide has a significant advantage. Something similar to what James' talk was on gallium nitride. Now, he's much higher power than we are. We're still in the three-volt battery stuff. We're not plugged into a wall. Anyway. Envelope tracking, thermal management in a phone. We're developing our fourth generation of envelope tracking.
The first two generations we're in production with today, one component carrier, two component carrier envelope tracking. We saw a 25% reduction in power dissipation using those technologies. We're sampling very heavily three component carriers, but also dual uplink. We'll actually add a second uplink. Again, we're estimating about 25% savings in current with the third generation. The big savings we believe to come is going to be in 5G. That very high peak to average that I mentioned previously, we really see a significant advantage for envelope tracking for 5G. We expect to be able to cut the current consumption by half by using envelope tracking. It's going to be very tough to do. I'll be clear. 100 megahertz bandwidth, it's very tough, but that we think we'll be sampling by end of this year.
We want to pull up a packaging slide to show you a little bit about what the modules look like, some of these integration modules. We started back in 2010. I might've been designing stuff back then, probably. Maybe not. It's a simple module. Had PA, switches, power control or a power management chip. Let me tell you guys, this was tough back then. I did it. I was doing it myself, bond wires. I tell you, the thing, it was tough. The standards have definitely evolved since then. Now we're looking at 100 CA combinations by 2016, adding filtering, switches, PAs, going flip chip. The density's significantly increased. By 2018, CA combinations have doubled again. MIMOs come in. We've had so much loss in these front ends, we had to add LNAs inside of the front end modules just to overcome the loss. Multiplexers.
We have so much density needed, we can't fit it all on the top side. Some of it's going to go on the bottom side of these modules going forward. I think Steve's going to go into that in his presentation. The density, it just increases, and keeps increasing in these RF front ends. To summarize, we feel like with our premier technology portfolio we have, we feel like we're number one or number two in all these categories, really positions Qorvo to really win the slots that we select to go after. With that, I'll turn it over to Steve Grant, Vice President of Operations. Thanks.
Good morning. You've heard presentations from my colleagues about all the great technologies and markets and opportunities, and I have the dubious distinction to be able to build all this stuff. It's been a very interesting time for me for the last handful of years working in RF, because I actually started up in silicon. I started in silicon for a long time and drove technology hard to the smallest nodes that are out there and then made a switch to RF. These wide devices, old equipment, and this is going to be easy, hopefully, you're getting a taste of just how complex this stuff really is.
My talk today, I'm going to be spending some time on talking about the core enabling technologies that we're using, I'll show you real things, the real stuff that we're doing, show you real data, hopefully get you excited about the work that's going on. I have a paid political announcement on capital efficiency and how with the technology work that we're doing, that we're making great inroads associated with just how much it's costing us to build factories and outfit the stuff we're doing. You can have great technology, and it doesn't matter at all if you can't deliver it in large volume, on time, with great quality and great cost. I'm going to end my presentation with some thoughts on that.
This is our roadmap, to the credit of the company, we've invested heavily in R&D along the way, as my friends and partners have already said, we have the broadest technology in the industry. It's a huge strength for us because what you see today and how parts are partitioned are not what it's going to be like tomorrow. What we're able to do is to work with our design community, to work with our system architects, to really put out there a broad array of parts that they can choose to optimize in the best possible way. Things that we thought would be happening three years ago are very different than what will be playing out today. This is probably the slide or the piece that gives me the greatest pride associated with the work that we've done.
Folks have talked about, we're either first or second in almost every category up there, we can build anything that we want to. We've made tremendous strides in BAW, the work is really being done behind the scene by very capable scientists, we have some of the elite scientists in this area, to drive the technology. We were behind. We've made substantial strides just in our base BAW technologies, but I'm also going to drill into where we're going and why we think we have a foot up as we go into higher and higher frequencies. BAW, for you folks, I'm sitting here I'm listening to all these acronyms that we're throwing down, everything. I'm like, gosh, do these folks really know this stuff? Can they really digest what we're talking about?
For BAW, it's really bulk acoustic wave filtering. Simplistically, what that means is that you filter the signals that are coming in through the bulk of the material. Instead of for SAW, which is surface acoustic wave, where it's on top. The wave actually propagates on the top of the material, that's the difference. Okay. For BAW does great work for about two gigahertz and up. Okay. You just look at the physics. I'll talk about the physics associated with it. It doesn't do very well at low frequencies. The devices get big. It has loss, all sort of stuff. For SAW, it's just kind of the opposite place, where it stops about two gigahertz. You might be able to extend it to 2.8 gigahertz. There's really a place for both technologies.
Some of our competitors have good capability in one and the other. As we've talked about here, when you look at the portfolio of what the parts need and what our customers are demanding, it's really broad-based, we have both, our technology's good. There, we were very surprised. This is one of the things I talked about before on where we thought the industry was going and what you needed. What's turned out is power amplifiers are becoming very important to us. We talked about SiGe and cheap silicon as kind of the going-forward play for the phones. What we're finding out, as everyone's talking about, it's so complicated. You have so many components in the chain from where the power amplifier is sending the signals out that you need massive power and efficiency to get it out.
We have done a lot of work both in GaAs for handsets, but also in GaN. We really see that this efficiency play, getting good linearity of the signal that's starting off and replicating it on the output is super critical for our parts. We continue to invest heavily in power amplifiers. I have a section on that. Todd hit on control signals also for our control technology for power for envelope tracking. That for SOI, we've worked heavily with an outside partner, with Todd's architects, with my process folks, to build a unique technology, a proprietary technology that we're only using. Okay. If you step away, what we're trying to do is we're trying to drive unique value for our customers and then have a broad range of technologies that we can integrate together. That's the strategy.
We press hard on R&D to always be a step ahead and advance the technology. What that allows us to do, which Eric hit on, is that we can look at what's out there. We can't do everything. I wish we could do everything, maybe several years down the road, but we can really go after the really hard, difficult technology challenges and provide value to the customer. That's our key strategy. With that, I want to throw this picture up. Eric talked about Phase 6, architecture for our open market, I hope you can see this, but what it's showing is that on these complex modules, we're putting a lot of components. Okay? It changes. Changes, more bands, everything else. This is for the open market.
We also make custom parts for premier handset vendors, the complexity is much higher than what we're showing here. This is the scale. Okay? The penny is 18, 19 millimeters across, and the parts that we're building, these fully integrated parts, are five or six millimeters on the side. Okay? Each one we talk about a SAW filter, you can see how small that they really are on this thing. I think that we just forget about how small the dimensions that we're talking about as we build and we try and integrate this. Remember, you have all the signals being routed at these very high frequencies, they talk to one another. There's huge magic associated with how you multiplex the parts on there, the routing associated, the shielding of the routing, and the fact that the complexity is just phenomenal.
It's not going down. I'm going to drill a little bit more into BAW. Talked a lot about BAW, Bulk Acoustic Wave technology. This is a roadmap that I showed last year. Once again, we're making great progress on what we talked about last time. Right now, we're converting from six to eight inch. We've been in eight-inch production for about a year. This summer will be about 30% or 40% of our products will be on eight inch. We've done great work associated with the matching of our yields, our performance, and see no issues associated with really moving forward on this technology. We're lucky we have two factories now. We have a factory in Richardson that's up and running on eight inch. We've qualified a factory in Farmers Branch and have ample capacity, really, to grow.
Worked on a technology called micro BAW, where we're shrinking the die sizes by, depending upon the design and stuff, between 20%-40%. Qualification activities are almost done on that, and that we should have everything done, locked up, in the third quarter, third calendar quarter this year. Great benefit for us. If you look at the physics associated with it, what you have to do to be able to get these frequencies is to super thin down the layers in order to resonate the frequencies that you want to keep and reject the rest. A simple example I give is like a tuning fork. You know how you have a tuning fork, and it resonates, and you hear that sound?
If you vary the pitch or the distance between the tines on the fork, you get a different sound. As you go up, that pitch and those dimensions have to get much, much smaller. That's at a macro scale or a micro scale, what we're doing is we're thinning down these layers to extremely thin layers. On my next slide, I'll talk a lot about the interesting things that happen when you have to thin down to be able to meet the resonance of these high-frequency signals. The other thing that I find very fun is that as we shrink these die, and these die are getting very, very small, what is becoming super expensive real estate in the wafer is actually the distance between the individual filters that are on there. Okay? Our filter sizes right now are in the 300-400 microns.
When I worked at Intel, the distance between the die that I built on microprocessors was 200 microns. Like that. There's a lot of science going associated with how can you go in and make that space between these dies as small as you possibly can, because it's a significant fraction of the wafer. We've been investing for a long time on our technologies to really shrink that distance, and there's some really cool stuff that we're doing there. The last thing I want to put on here is that it's not stopping. We don't have the luxury of going from node to node, but we look at things associated with 3D stacking, ways that we can move routing, and that we're working on something on stacking resonators. We're not done, we're working on it, but it's there.
We'll continue with our scientists really to pursue and to push, continuing this cost and this performance vector. Now I want to talk a little bit about SMR. This gets kind of geeky, sorry. SMR and FBAR. Those are two ways of producing a bulk acoustic wave filter. We use a technology called SMR, which is a solidly mounted resonator. Okay? How these goofy things work is that simplistically, you have two electrodes or contacts that take the information in and out of a material that really vibrates and gets a signal called a piezoelectric. Okay? The devices that we build have the two electrics. They have the piezoelectric, and then there's a thing called a Bragg reflector that actually reflects the energy back up into the piezoelectrics. That's our device. Okay?
What FBAR does, very good technology, very similar associated with the electrodes. They have a piezoelectric in, but they have an air cavity underneath that allows it to resonate. People have written, you look at the physics, it's a good structure associated with the coupling that occurs. Okay. The challenges that this comes as you go to higher and higher frequency, is that the losses that are generated in these devices because they're so thin, starts to be impacted heavily by the electrodes that you're using. The electrodes have to have two things: they have to be acoustically very good, but electrically very good. As you thin the layers, it gets super hard to get both. We've invested the time and energy associated with our electrode technology, and I think it's unique. That's one takeaway. Layers are getting super thin.
The precision you need and the materials you work and the material structure is very important. The second one that's really interesting is heat. Okay. What you're doing right now is, we talked about having to pump a lot of energy through these small devices, and we're pumping through about a watt. A watt's about the max amount, and there's loss that occurs in there, and it's just like a resistive element, and it heats up. Okay, this is a big challenge, right? People don't think about this stuff, right? This is a big challenge for us. Inherently, because of our structure has these reflective layers in it, big silicon base, no air, we can get the heat out. Okay.
We can get the heat out of this thing where there's less conductive pass, and we've done simulation associated with industry standard tools, and we see a very big difference between the inherent capability of an SMR BAW and an FBAR. People are smart. Everyone has scientists. We just look at the inherent advantage as we see it right now, and we think that we have a really good opportunity going forward. That's nice, right? We're measuring the stuff. This is real data. Okay. This is real data on our part at 5.2 gigahertz. How this thing works is, I want to show these charts, what they mean. The left one is, I want to send a signal through a filter, and it says how much loss did I have.
What you want is to have very little loss, and you want it to be up. You don't want to lose any energy through it, and there's industry standards associated with this. Our benchmarks across that wide bandwidth were good. Okay. The other thing we showed on here is, even though you go in and you heat it up or cool it down, we still have margin associated. Because remember, the materials are very thin. Winter, summer, still has to work. Okay. The second piece is rejection. What you want is you want all the signals that are outside the area you want to pass through, you want to get rid of it. What's important is the left and the far right side. These are prototypes. We'll make it better.
We're excited, and hopefully, that's coming across associated with the work that we're doing on 5G and the technologies that we have. I want to talk a little bit about power amplifier technology, just a little bit. Big deals for us are increasing the power out of our PAs. It's to make them more linear so that they replicate the signals very well that are coming in. The other big thing is ruggedness, that as there's mismatch in the antenna, that it can withstand the feedback to the power amplifier, and as has been discussed, we have really good technology here. Continue to invest heavily with next generations that improve both linearity, ruggedness for us. The other piece we're pushing is we're pushing these very, very fast frequencies that Todd talked about for the 28, 32 gigahertz. I love James to death.
He makes my head hurt associated with how fast he wants to go, and his device is now they go up to 100 gigahertz per GaN. The technology's cool. The message going here is there is investment. We continue to push very, very hard to try and advance our technology portfolio. You got to assemble all this stuff. You got to put it all together. Forever, what we've been able to do is to go in there and crunch the devices smaller and smaller and smaller together, and what people have figured out is that's just not going to scale. What we're doing right now is we're starting to stack both sides, which is pretty cool technology.
The sizes for our most complex modules were around 90 millimeters squared, and the new things we're looking at are about 55 or 60 millimeters squared. Technology's hard. Customers really value performance, but they also really value size. I put this one in for Mark, our CFO, we've been lucky enough that we've invested a lot in factory capacity, but really just wanted to reinforce that with the technology advancements that we're making, our cost of capital per every die or every wafer build is going down substantially. Using industry standard tricks to go from the mixed wafer size, we talked about six to eight on GaN, we're going four to six, on SAW we went four to six. Where we can, we'll continue to push that. Big deal for us, right?
You get 1.8 times more die per wafer, and it costs you about 30% more to build it. We talked about micro BAW. Micro BAW for the same exact capital bill gets us about 20%-40%, depending upon the die size. I also tried to elaborate on this, see how important the width was between the die, and that should give us another 10%-15%. We will continue to look at these opportunities to drive down the cost of capital. I talked a lot about the technology roadmap and super cool, and we're really excited, but it doesn't mean anything if you can't build it. What we've done on this vein is we have fantastic research scientists to help with these very creative ideas.
We've gone to the industry, we've found people that are professionals at developing technologies, mostly from the silicon industry. We have large teams that take these ideas from our research scientists, call them out, make it manufacturable, rigorous process, then start the manufacturing ramp. We have a very large development organization just dedicated to get technologies ready. Very great thing to do. We've also invested heavily in inline monitoring, both inspection, automated inspection, electrical piece. The thing that we've done that's pretty cool, that's just like everyone else is doing this, well, leading companies, Samsung, Intel, those folks, is doing a lot of diagnostics of the tools. You're monitoring everything on the tool so you know if there's changes in pressure or flow rates or anything else, if there's any problems, you turn it off.
The cool thing that we're doing, too, is you take all that data of what's coming off the tool, the electrical, the inspection, put it all in a database, then you run algorithms, you call it AI, deep learning, whatever you want to call it, to find problems. It's pretty cool, we're just evolving on this thing, this is the type of buyer we wanted. It's we find problems immediately and fix it. The other big thing that we're doing is that before we even launch products, we're stressing the heck out of our products and trying to find defects before it gets to the customer. The philosophy is we want to find problems. Let's find them early, let's beat them up, let's make sure we can fix them before we go. Once again, nice words. Here's the data. Data analysis.
Some is anecdotal, some is real data, we're getting a lot of very good feedback from our customers. Huawei just gave us an award the last 2 weeks as the only RF supplier for their premium or best quality team award. Stuff is working. We're getting scores from our mobile partners. We're 1st. Not all of our partners will allow us to share information. We're getting great feedback associated with the ones that are willing to share and say that ranked against the other RF manufacturers, this is where you stand. We do this process called clean launch, which takes all those ingredients that I talked about, put them together, make sure we have a scorecard of how we burn in parts, how we measure yields, everything else in order to do it. It's working.
The chart on your left just shows the results, this is from very large customer and shows the data rates of the defects per million that we're seeing. It's kind of interesting for me, right? Been in this industry for a long time. Remember, this is the module-level defect level. We're like 10-50. Remember, there's 30 components on it. We're like one defect per component per million. Can't tell you what it was 30 years ago when I was working in silicon. Huge benefit. Okay. That's my talk. Hopefully, what you took away from it is strength of technology portfolio, continuing investment in leading-edge technologies, opportunity really to partition and use them in the right places.
The walkway, hopefully, on making money is our development cycle, our monitoring systems, our pre-launch activities, and hopefully satisfying our customers to the utmost. With that, I'll turn it over to Mark. Thank you.
Thanks all of you for joining us today. Just for some introductory comments. Today we've covered how Qorvo is a leader in the technologies and products that enable wireless connectivity. With the progress we've made on shaping the portfolio, driving operational excellence, and reducing capital intensity, Qorvo's in a great position to serve our markets and profitably grow. Today, what I'm going to cover is how this shapes our outlook and our capital allocation. These changes that we've made are giving us great confidence that we can achieve the results we've laid out. Nothing has really changed on this slide from last year, in the sense that we've committed to deliver above-market growth, expand margins, and drive free cash flow. Our outlook for growth is positive, with strong organic growth leveraging technologies and positions in advanced 4G, 5G, IoT, defense.
On margins, we've turned a corner on utilization. Productivity programs are yielding real benefits, we are laser-focused on portfolio management, as you heard today. In fact, we are on track to the operating margin targets at last year's Investor Day of 33% by fiscal year 2020. Finally, this management team is committed to free cash flow growth through profitable sales growth and lower capital intensity. This time last year, we committed to doubling free cash flow in fiscal year 2018, which we did. In fact, we did better than that. We did two and a half times from fiscal year 2017 to 2018. Our free cash flow margin has gone from a low of 7% in fiscal year 2017 to a forecast of over 20% of sales in fiscal year 2019. We are absolutely making progress. Our growth prospects are excellent.
IDP enjoys differentiated technology and compound semiconductors and systems integration, helping provide market leadership in defense, IoT, and 5G. In mobile, trends of greater RF complexity and integration play to our device design and production capabilities. We're seeing this play out in a number of areas where performance is key, such as Phase 6 architectures and in the mid and high bands, including sub-6 gigahertz 5G. Our gross margin outlook remains unchanged, with a return to over 50% in the second half of this year. We've been there before, as you can see from this slide, and we expect to be there again soon, driven by a mix shift to a more profitable portfolio and as utilization improves from last year's lows.
Active portfolio management, which you heard Eric and James talk about, and CapEx discipline, which Steve mentioned, give us confidence that we can expand our gross margin further. You see that in the fiscal year 2020 highlight. Likewise, we continue to improve our OpEx efficiency. R&D spend is increasing, importantly, it has become focused on our most differentiated products. SG&A is also trending down as a percent of sales as the company continues its lean journey and leverages a single SAP instance, which we completed this year. With our technology breadth and market opportunities, I don't see Qorvo operating at the lowest industry OpEx level. What I do see is us continuing to better leverage spend as a culture of continuous improvement drives effectiveness and efficiency. Progress on our operating model and CapEx discipline is yielding stronger free cash flow.
From our low in fiscal year 2017, we've come a long way. In fact, over each of the last two full quarters, we generated more free cash flow than in all of fiscal year 2017. This fiscal year, we expect to generate at least $700 million, but are targeting $800 million. With growth in the right areas, continued operational discipline and productivity, and lower capital intensity, we expect free cash flow growth to continue. Growing cash flows and a strong balance sheet provide us the flexibility to reinvest in our business, selectively acquire where technologies and businesses fit into our objectives and our model, and return cash to shareholders. In fact, today, given our cash flow outlook and strong balance sheet, we announced a new share repurchase authorization of $1 billion to sustain capital returns to shareholders. Finally, our model remains unchanged, and we're approaching these goals quickly.
In fact we'll soon need to update James' model, and mobile actually isn't too far behind. We expect to grow above market through technology leadership and positions in 4G, 5G, defense, IoT, and broader connectivity markets. We see gross margin above 50% through growth in the right areas, productivity, and capital discipline. Our OpEx should continue to trend down without compromising the ability to pursue quality growth. As I mentioned earlier, we are still targeting operating margin to be around 33% in fiscal year 2020. Thank you all for your time today, and now the rest of the management team will join me as we move to Q&A.
Well done.
Well done.
Yeah.
That was fun.
Oh, yeah.
All right. Who would like to volunteer for the first question? I'm looking to my right. Go ahead, Kimberly. Ladies first, sorry, Ed. Thank you for joining us. Go ahead with your question, please.
Of course. Sure. The first question I'd like to ask is, the investment made I thought this was
Sure. I'm sure Eric would love to answer that. Thank you for staying away from the marquee phones. Eric?
Yeah. Thanks for the question. As you said, we are seeing content growth really across all tiers. The content in the mid-tier, what we call a performance tier, last year as we talked about the Band 13 quadplexers coming out, that was a bit of a headwind. This year, those are coming back in, which gives us a tailwind again that we had two years ago with more carrier aggregation driving more quadplexers in those mid-tier handsets. Going forward with Phase 6 integration, you're seeing, if you saw there was 12 BAW filters, 6 SAW or 4 SAW, what was it? 8 SAW die, in that total combination. A lot more filtering coming in to cover more bands, and those are all going to that sort of performance tier.
Not the flagship tier, but that mid-tier where there's hundreds of millions of units per year being shipped.
Perfect. Thanks. Eric, I guess this next one as well. We saw that slide about the rollout of 5G in smartphones and radios and whatnot, and it'll first be on a conjunction with 4G and it will be on a standalone basis, and then it will move to a standalone basis. When 5G does move to a standalone basis, will it still be a cumulative effect in terms of you'll still have that 4G, 3G content that we're seeing now, or will it only be 5G content?
Yeah, that's an excellent question. I should've probably qualified better in my remarks. When we say standalone 5G, it just means that the 5G radio signal is able to operate on its own network independent of a 4G network. The phones themselves still will have backward compatibility, just as 4G phones today still have 3G and 2G. The 5G phones will still have 4G, 3G, and in some cases, still 2G even in those handsets. The 5G signal will have its own dedicated network and complete with all of the signaling and all of that. It's a different protocol that's independent when it's standalone.
Perfect. Thank you.
Thank you. We'll go over here.
Thanks. One for Eric. There's a lot of growth in mid-tier, but most of all the RF TAM is in the high-end and most of your growth and your concentration of revenue's in the high-end too. Given the bumpy road that premium phones have had over the last three years or so, the fact that they're redesigning, all those guys are redesigning their front end every year. They're throwing a ton of. This is a big benefit to you. What makes you think that they're gonna keep doing that? I know the roadmaps stretch out three years and that's what it looks like, but if you're not selling a lot of these phones and all this investment in the RFEE isn't really turning into unit sales, why keep up doing it?
Do you expect maybe at least like the tick-tock method of Intel is gonna stretch out a little bit more? Then I'll just give James his now so he can think about it. The GaN business is rocking. That works out really well. You buy your wafers from another wafer supplier. I'm sure some of that comes from Cree. Cree just bought Infineon's RF business, which gives them finally a channel into that market. This is kind of repeated my question on a conference call, so a little different take here. Given their aspirations now to stay in the game and to actually expand it, why wouldn't they enter the military business, which is who your core GaN is? Is that something you're looking at? Is it something you worry about?
If your supplier starts competing with you head on in your core business, what does that say for the dynamic in terms of getting wafers, et cetera? Then one for Steve, and I'm done.
I'll go first.
You go first.
First of all, you're right. A large part of our TAM and growth is in that flagship tier, but there's actually, we have a very large business in the performance tier, and that's growing dramatically as well right now with the Band 41 quads coming in. We mentioned also a lot more tuning and antenna flexing coming into that tier as we covered. There's an awful lot of growth there as well. To answer your direct question on the flagship tier, why do they keep doing new radios every year? There is an absolute continuous march to add more data connectivity and higher data rates in these devices. Applications where we provide the ability to get the data in and out, applications are waiting to be developed to take advantage of.
Augmented reality is just right around the corner waiting for better connectivity in these solutions, and that can actually generate a lot more handset sales. There will be cycles of major upgrades in handsets based on apps usually, but those apps are dependent upon new hardware and faster data rates in particular. We don't see anyone taking the foot off the gas at enabling it.
Ed, for GaN, just like all of our other internal semiconductor processes, we have a very robust supply chain. We don't rely on one supplier. We have multiple suppliers across all the nodes that we have in GaN today. Somewhat, we're certainly isolated from any kind of effect that you've discussed earlier. As far as the defense market or GaN in general, I think we are set up now for really three competitors that have the full integrated value chain. They've got the wafer fab integration, packaging and test, and the channel to market. We're certainly one of those three, happy to be one of those three. I think we'll get certainly at least our fair share, if not more than that, in all of those spaces.
As far as defense, Cree's been in the defense business in GaN for the long time, since the '90s, I would suspect, or early 2000s along with us. They've been in that market consistently.
Final question for Steve. Three years ago, you were launching into a new program at one of your largest customers in TC SAW, Chelmsford, Mass. You just moved from 4 to 6 inch wafers. Now this year, you're about to launch into another big program. This time in BAW, Richardson, instead of SAW, Florida. After you were awarded the design win, if we go back two or three years, after you were design win, but before you started production, what were the generally the yields in TC SAW as you entered into that program, and how does that contrast to where you are today in Richardson, Texas? Because you're on 6-inch there, you're going to 8-inch. Are we looking at anything similar? Thanks.
I'm smiling lots. It's a great question. I'm not going to give you the specific yield numbers, but the yield numbers on TC SAW were 60%, 70% of what they were, the 6-inch versus 4. Right now, between 6-inch and 8-inch, it's about the same.
Are you saying that in your BAW business now you're about the same yield you were heading in on TC SAW?
No. My 6-inch BAW yields and my 8-inch BAW yields are identical.
Okay.
Okay?
You're saying the TC-SAW was substantially lower going into that production then?
That is correct.
Okay.
That is correct.
Thanks.
Thanks for the clarification.
We'll go over here.
Thanks. Two questions for Mark and maybe Bob as well. Just on the gross margin, you showed low 50s in Q4, and you look at the implied guidance for fiscal 2020, it kind of sustains that level for the full year. Two variables, I guess, I'm curious as your visibility this year into the volumes to support those utilizations, and then just if any perspective on where those BAW parts are in terms of a yield curve and any kind of timing and milestones to kind of guarantee that you can get to those levels.
For this year?
For this year, yeah.
Yeah. We are in a large part of our business is the consumer product, so the forecast there can change. Our view right now is solid, or we wouldn't give the guidance. Volumes support the utilization, which you see, which obviously supports the margin. Importantly, it's a utilization and a mix profile. We also see the mix turning more favorable in the back half of the year. As far as operationally, as Steve mentioned, yields are very high in Richardson. It's an excellent fab, and a lot of confidence with the customer at the operations of that fab. We're not as concerned as we are excited about the way things are playing out here. As you saw Eric's view of where we're spending our investment dollars, where we're spending our capital dollars, it's in that area.
When you see the architecture changes coming, particularly the Phase 6, which shows it blowing out for the rest of the market, that's good for us.
Perfect lead-in for the second question. You talked about free cash flow of at least $700, most likely $800. Can you kind of just talk about what gets you between the bottom end and the high end of that range? You kind of front-loaded CapEx last year. You get some benefit here. You're also doing things like the 8-inch and strength. Can you just talk about, you're probably not going to guide CapEx for next year, but just how you think about supporting this fall ramp these lead times.
That's a good question. Just a clarification to your comment, I did not say likely $800. I said at least $700 and targeting $800, I believe. In any case.
I know the number you're giving us, though. It's $800. That's what he told us to go do.
Nice try.
We'll work hard at that.
I know what my goal is. In any case, back to the volumes, it's a number based on a sensitivity around what could happen. There's a lot of complexity in cash flow forecasting, obviously, because it's a function of some things that are good for your business may actually have us improve free cash, and some things that are bad for our business, vice versa. It's a difficult number to forecast. Going to be a function of do volumes track where we think? Do the productivity programs yield what we think? But right now, I'm comfortable with that range. Normally, when I give a range, it's a sense of highly comfortable in the low to midpoint, and comfortable, but less so as you go up.
As far as CapEx going forward, I think I've made the comment before that really I don't see certainly a forecast horizon that we look at, which in this case is through fiscal year 2022. I don't see us ever spending at the levels that we did in 2017. I think we spent 18% of sales on CapEx in fiscal year 2017. Fiscal year 2018, we were down below 10, I think nine. This year will be about the same level on a dollar basis, going forward, it'll be, best we can tell now, about that dollar level or maybe even a bit lower. It's a function of the ability to just be more thoughtful about where we're spending our CapEx dollars. It really gets back to this portfolio management question.
We're not going to build another SAW fab, for example, based on the current plans, but we may choose to expand in BAW when needed. That brings up another point about CapEx, is just the efficiency in which our incremental spend is occurring. Steve mentioned the wafer expansions, the die shrinks, the various other programs and productivity. A lot of those, for example, in BAW, most of our fixed asset base now is 6 inch. We have a lot of room to very efficiently expand there. As Steve said, you combine that with the die shrink, the dicing efforts, you're really becoming more efficient.
Thanks. Go over here.
Hi, thanks. Thanks for the presentation. A couple for on mobile. For Eric or Todd, as your customers and you are drawing up the 5G designs, should we expect the architectures to develop with a discrete 5G PAD next to, say, a mid-high band PAD from you on the 4G side, or do we go straight to an integrated module, and where are you positioned?
Todd, do you want to take that?
Sure. We're seeing requests for, I guess, discrete PAs with possibly filters external. We're also seeing requests for an integrated type PAD where it would sit next to maybe the other PAD, a PA plus a filter as a PAD. We're seeing a few different requests coming in. Again, people are still architecting the radio still. They're not locked yet in the industry, they're keeping their options open as the standards evolve here.
What % of the phones have moved to 4x4 MIMO already, and what is your outlook for that over the next couple of years? What's your content increase versus a two antenna phone? Todd's presentation talked about seven to eight antennas. What does that do to your content?
A very small percentage of phones today are actually 4x4 MIMO. It's really just getting started and kind of demonstrating capability at this point. We do see it expanding over the next couple of years, but it'll still be limited to a relatively small portion of the market. It won't go to 100% of the market, for example. That's your true LTE-Advanced Pro. It's the flagship tier of the market. You can model how big that is to get an idea of the penetration. We don't break out content gains by specific functions like that because it's rare that you would add 4x4 MIMO without adding other bands and other things that are complexity around it. We don't break out value per function to that degree. Okay, we'll go over here.
Thank you. Srini Pajjuri from Macquarie. Again, thanks for the presentations. I have a couple of questions on the BAW side. Again, going back to the 5G team, I guess one of the reasons we saw so much content increase in 4G was the bands were pretty close together, and the guard band was very narrow. I just want to understand how 5G bands are, how do you see them, and if you see the similar situation in terms of the complexity and the 5G bands being next to each other in different regions of the world. Because one of the things, at least one of your competitors claims that it's mostly TD, it's not FD, and as a result, you don't need BAW filters. You can get away with a TD filter. Then I've read somewhere that you can actually get away with a CMOS filter.
You don't even need a fancy BAW filter. I want to hear your thoughts on that. Then, second question, I'll go ahead and ask it. In terms of the DRX, today most of the DRX modules use SAW filters, and your focus is on BAW. I'm just curious what your approach is gonna be as you enter that market. Thank you.
Todd, you wanna take the first one? I'll take the second one.
Sure. Yeah, they are correct. The new bands of 5G are TDD. You don't have a TX and RX on at the same time. That's more like an FDD system. The TDD systems, the issue you have with the TDD system in those bands is coexistence. I think you looked at my presentation, I talked about coexisting with Wi-Fi. Wi-Fi is very close, especially Band 79, you have to coexist. You also see some coexistence issues with some of the other LTE bands at the lower frequencies. It depends on The call of the distance case on how much BAW activity needs to happen. We expect the BAW to be definitely in the '79 and possibly in '78. It's just going to depend on how much coverage these phones have.
You could make a case that maybe if they're a regional skew phone, that you might not need as much BAW content as you could in a world phone. There could be some use cases.
You also do agree at 600 megahertz, we will be using a different technology.
Yeah. You go to the other end of the spectrum. 5G is not just being rolled out in the 3.5 gig, 4.5 gig bands. Good point, Bob. It's also coming down to 600 megahertz, which requires yet a new TC SAW filter technologies that we've developed. It's also retrofitting with some of the other LP bands, Sprint's drive in the 2.5 gig area, to where they're going to start bringing up 5G into that band. 5G's going to go across that whole spectrum over time from 0.6 probably up to 6 gigahertz.
Maybe one more thing to add to that. That discussion, as well as most of the discussion you're referring to, is just about the main radio path. It's still ignoring all the antenna complexity out there. Unless you're going to add half a dozen more antennas somewhere for the 5G stuff, you're going to need to share it with the 4G antennas and the Wi-Fi antennas and the GPS antennas and so forth. That antenna plexing is certainly the work of BAW, that's able to do that at very low loss fashion. That will still come into 5G. Now to your second question about the diversity receive, similar answer in a way. As more carrier aggregation and more MIMO is coming in, the loss gets higher. When you're trying to address these with SAW filters, the loss really adds up quickly.
BAW can come in and bring differentiated, especially in the mid and high band, bring much lower loss to the diversity function. The reason we're entering when we are is because we're seeing the complexity and diversity go up to the point that the losses with SAW just are no longer tolerable.
Thank you.
Thanks. Chris Caso from Raymond James. Mark, just wanted to dig in a little bit on the 33% operating margin target for next year. Just a little bit of an explanation of what it will take to get there, how much is within your control, how much is the market, and I'm sure you probably don't want to provide revenue guidance for next year at this point, but I assume that requires some level of revenue growth. Maybe give us an idea of how aggressive those assumptions might be as you look into next year.
Yeah. Chris, good question. I don't want to go too far into giving granularity about a single number target, which we still believe are on track, but it'll be more of the same. It'll be investing in the right areas, which we think will grow at or above market. Of course, if the market drops for consumer handset reasons or other things, obviously our volume could drop as well. We believe we're in product segments that we will be able to grow above market. The mix shift and the sustaining high levels utilization will help with the gross margin, and we should see continued expansion there. As we've tried to make clear on the OpEx side, between more efficient R&D spend, and just more efficient overhead spend, we're gonna continue to get leverage there.
We have a clear line of sight, but in this business, of course, there are always some things out of your control. Hopefully we've got enough opportunities beyond our own target to offset those.
Okay. Just as a follow-up, in some Analyst Days in the past, you guys have quantified kind of revenue opportunity per device. I wonder if you look into 5G, and I guess there's a couple of stages of that, non-standalone versus standalone. Where do those revenue opportunities come in for you? Is there a way for you guys to quantify the dollars of opportunity that you have in front of you when those things hit?
Yeah. You're right. Typically, we show the various, like an entry handset and a premium and a performance and try to show dollar content. The reason we haven't done that this year, in particular, didn't include 5G, is because there is much more of a continuum now, and it gets really hard to classify which handset goes here versus there. Also the content per device is part of the story, but the mix between them is another big part of the story. That chart kind of has outlived its usefulness, maybe. When you look into 5G, we show from the TAM, $1 billion of TAM beginning in CY 2020. That's probably your best guide at kind of quantifying how much 5G content we see coming. To give specific guidance of specific handsets, it's obviously going to enter at the high tier.
We can't look too forward in terms of giving our customers production plans for volume devices. Calendar 2020, big content in 5G based on our TAM model.
Right. Let me clarify that. Is it more of kind of gradual up to 2020, and then 2020 is when you'd expect perhaps more of a content step change, or is it even smoother as you go into 2020?
Yeah, you can go back and look at that TAM chart and see it. It's more of a step function change in 2020. There's some in 2019, but it really begins to go much bigger in 2020 with a full year of phone shipping.
Okay.
Hi, Karl Ackerman from Cowen. Earlier today, you gave a pretty robust outlook on your IoT applications across various standards, while also talking about improving, I think, OpEx discipline. I'm curious how you prioritize R&D across your IoT portfolio and perhaps where you expect to receive the most design wins over the next 12 to 24 months.
I'm gonna start and you can finish.
The first way is by this multi-protocol chip. If you look at what we've done, we've been able to address many opportunities with really one R&D effort in products that were released last year. I'll let Cees talk a little bit more about how we pick and choose markets, if we want to talk about that in all the verticals in IoT.
Now, as you saw in the movie, it gives a little bit of an impression that everything is going to talk with everything else. That's a kind of a dream that we think over time will be the endpoint, the way the market is going there is by certain verticals that create extra value for consumers or in businesses that kind of pull ahead. We are very strongly positioned at this moment in, for instance, the remote control market. We are also positioned well in several other opportunities in the home. There are also opportunities outside of the home, more in the industrial space or in the retail space that we are pursuing that essentially leverage the same benefit, getting more data available faster, and therefore can be turned via artificial intelligence in good decision-making for business owners, et cetera.
That's the way we look at the market. Over time, we expect convergence, at this stage it's really being very selective and winning in the markets that we select.
We are also very focused on the connected car. We see significant content capability in a car, anywhere from LTE to our C-V2X capability. That's another focus area for IoT for us, to make sure we capture that content in the car. Maybe if I could just add quickly on the cellular IoT. It leverages very well all the activity we're doing now. We're able to reuse, for the most part, the investments in the handset to go into those cellular IoT markets.
Just kind of a follow up on a, I guess, kind of question for Steve. Steve, you talked at length about improving your wafer-on-wafer capital efficiency and kind of gave a roadmap across performance and the cost of modules. I was curious if you may quantify the cost savings you expect to achieve as we talk about the shift from micro wafer-on-wafer and in die-sizing over the next year or two.
I don't want to give a firm number. I'll give you ranges associated with it. The math is easy for 6-inch to 8-inch conversion. That you can do, and the number I put out there was about 30%-35% more for wafer cost in total to do that. The piece on micro wafer-on-wafer is dependent upon the design, how much you're able to get, and that also plays into the loading. The numbers we're seeing are between 20 and 40-ish type numbers. Once again, the reduction within is also dependent upon the die that are there. You can kind of see, and we try to make a range associated with each one of those to try and give you a feel. To give you an exact number would be just wrong.
Hey, guys. Harsh Kumar, Piper Jaffray. Question for Bob. Bob, in the silicon world, we're seeing ASP compression get smaller or lower. You guys are at three to four or five legitimate players in your industry. Can you talk about the ASP trends you guys are seeing, let's say apples to apples, or let's say oranges to oranges products in this case? What will it take for your margins to go up from here just as the silicon guys are seeing them go up because of lower pricing?
First of all, Harsh, the good news in our industry is every year they're asking for more, and if you look at the roadmaps, when we just take the example of the pictures of Phase 6, our ASPs are actually, in Eric's business, going up over time because we continue to integrate more and more and add more functions and more features. That's one thing to keep in mind. The second thing is, as Steve's already pointed out, we've been working hard to drive our costs down through diameter changes and die shrink. We've actually got something interesting going on with the businesses today. From that perspective, we feel pretty good because, again, to make room, you heard both Todd and Eric talk about the RF section, they don't give us more space, but expect us to put more in it.
We have to make room for 5G, which is why a lot of these customers are looking at Phase 6 and already saying, "Geez, I got to make more room." We're actually getting paid for that integration as well.
Question for James. One of your competitors, MACOM in specific, talks about GaN-on-silicon revenues this quarter, production, commercial stuff. I'm curious about your views whether that can be a disrupting technology, particularly on the cost side, or it's just some tangential play.
The majority of the market's still GaN-on-silicon carbide. It offers significant advantages that we've talked about in the past with heat removal, and therefore better efficiency, better performance. I think that story still holds true. We certainly pay attention to GaN-on-silicon, and it's coming up in some commercial foundries, and we'll continue to sort of pay attention to that. There will be some slots, probably in fairly low power levels, where that technology will have a home and play. I think the bulk of the high power, if you think about replacing LDMOS, will be in silicon carbide at those high power levels.
If I can ask one last one maybe for Todd. Todd, in terms of this mid- high PAD, is this a 5G future-proof product, or will it change all over again once we get to that technology? Secondly, why did the big architectural change happen? What was it driving for specifically?
Okay. I don't know if I follow the second part of the question. If I can answer the first part. The MHB, at least in the architectures that my system guys have drawn up, we don't see, at least for the new radio bands, the 3.5, 4.5 gigahertz, really driving much change in the MHB architecture. We think they'll be independent paths for the most part. We don't see a lot of overlap there in those two PADs, for instance.
Okay. Let me clarify the second part of the question. I guess, why did the industry, or why are the top tier guys trying to move to the mid-high PADs?
Oh, yeah. It has to do with carrier aggregation. I think we've talked about it a bit in the past, but in order for you to have two mid bands coming in for the CA modes, the filters have to be multiplexed together. You have two mid bands coming in. The filters have to be hanging off the same node. That's what we call multiplex. It has to be all integrated into that one PAD, and that's the mid band. The carrier aggregation is from the mid and high bands now, to where you have a mid band plus a high band. You might have a Band 1, which is 2.1 gigahertz plus the lower mid bands together, like 1.7, so it's a mid-high band. For same reason, they get multiplexed together, the filters do. The loading is they see each other filter.
It's all to do with the multiplexing of those bands for CA modes.
Thanks. We'll go over here.
Great. Thanks. Just a question for Eric. A year ago, there was a lot of optimism around that mid-high band program, and it looks well-placed in terms of some of the initial traction. Can you just give us an update on how that's progressing, maybe at a high level in terms of that opportunity set on the premier side, how it expands? The second question would be on how you see the China market, the case for integrated products versus discrete approaches.
Yeah. We couldn't be more happy with the progress we're making on these programs, both in the flagship tier, we've got multiple customer sampling now and expect to be in production soon. Also in that mid-tier, if you will, or the performance tier, as I mentioned. Technology's doing great, fab is doing a fantastic job, design teams are executing. Everything is just going very well there, and that's a lot of our optimism. When it comes down to how much penetration, I think the second half of your question is how deep could that go in the portfolio? That's going to be interesting to watch. We're the first to market with this Phase 6 solution that has all this level of integration. It is more costly for the customer to pay for this integration.
Even in that performance tier, our customers, which are mostly our large Chinese OEMs, which are by now getting to be quite large, in fact, in their volumes, they've all got their eye on 5G as well. Their motivation is to make 4G as small as possible, put it in the corner someplace, so they can have options for rolling out different 5G and other content they're looking to. It's going to be a typical kind of price versus volume curve, but we're going to make sure we're getting paid for that integration as we bring it to the market. We'll see how it penetrates down.
Great. Thanks for the call. Maybe just a follow-up question for Mark. When we think about the buyback announcement today and how you approach cash return versus acquisition opportunities.
I'd like to do both. We're continuously looking at acquisition opportunities, as we've said it many times in the past, principally for technology purchases for Eric's business, and you can see some of those disclosed. Both technology and business acquisitions for James' business bolt-ons. GreenPeak Technologies, you saw a case today, that's been very successful to us in gaining market exposure in an important market. On capital returns, we're committed to return cash to shareholders. When you look at our cash flow profile and how it's been improving and what we expect going forward, you look at our balance sheet, which has become very conservative, we've certainly got the capacity to do what we want on acquisitions and continuing to return cash to shareholders. Hence the billion-dollar program announced today.
Great. Wendy.
First, a quick clarification for Mark. I understand it's just a target, but the 33% op margin target for fiscal 2020, is that a fiscal year target, or do you just expect hit 33% in one quarter?
That's fiscal year target. Fiscal year.
Okay, great. Then two for Eric first. Eric, if you look at your TAM, the RF TAM that you put up, it looks like 4G content beyond 2019 is relatively flat. All the growth comes from 5G. Are you at all worried about a delay in 5G and what that might do to the TAM?
Yeah. That's a good question. The way we're measuring that is the 5G, what we're putting in the 5G bucket is the additional value of content for 5G. If it does push out, a lot of that value will switch back to 4G. There'll be even more LTE-Advanced Pro handsets, for example, shipped to replace the ones that are currently slated for 5G. I think if it does push out, you'll see it replaced with LTE-Advanced Pro. Frankly, right now, it's looking like if anything, people are trying to pull it in, similar to what James has said.
Yeah.
It doesn't seem likely, at least right now, there'll be much, if any, delay in 5G. It's a question of how quickly we can bring it to market.
We're definitely seeing on the infrastructure side no slowdown. It would appear that that's the capabilities coming online about how it's been projected across the industry.
A 5G question, I think, follow up to Srini's question. 5G bands tend to be fewer in number and much wider in bandwidth. I guess, as you look at the dollar content for a 5G band versus a 4G band, do you have higher content in those 5G bands to sort of make up for the smaller number of bands you might incorporate in the phone? If there is greater dollar content, does that come through the PAs, does it come through the filters, or does it come through some of the new devices like DRS modules and antenna flexors that you need to support that complexity?
Yeah. It's a complex answer, unfortunately. That's why we showed the block diagram on my section, which showed that when you add a 5G component here, it actually affects a lot of the other 4G stuff too, because they have to coexist with each other, right? The 4G stuff can now not interfere with that 5G player that's over there. It's really very interrelated. We don't even look at what that particular 5G component itself does. It's part of the entire system.
It's more of a complexity rather than a number of components.
It really is. Yeah. Adding that one 5G component affects an awful lot of the 4G stuff around it, that's why there is additional 5G bump even in the 4G content because of that. Yeah.
Thank you. Amir?
Thanks so much. David Wong, Wells Fargo. Can you give us a rough idea of the economics behind your mobile growth? Do handset ASPs go up? Do you expect to pick up unit growth? Are there component areas where handset makers can save if they have to pay more on the RF side?
Yeah. It's a good question. We're certainly not counting on handset unit growth. Not at all. That's not part of our equation. It is, as we talked about, the content in each generation is getting higher year-over-year. How they afford that is up to them, whether they raise their prices or take the cost out elsewhere or take lower margin. I can't answer how the handset guys are going to necessarily deal with that. I will say that we're on average 6%-9% of the BOM, depending upon the tier. We don't crest 10% of the BOM in any handset that I'm aware of. Just to put it in perspective. We have room to grow before it becomes a terribly pressing issue for the handset provider.
We think generally handset ASPs as well are going to be going up to cover this and memory and the fact that they bring more value, frankly, going forward.
Thanks.
Thank you.
Ed Snyder, Charter Equity Research. Just one more round. Eric, most of your growth in mobile has been in the high-end. Most, really, actually, really top-tier customers. You had a little bump with China there. That seems to be shifting now because of the unit volume issues there, but also because China announced in the summer of last year that they want to get into 5G. Aren't they pretty far behind? Aren't most of the Chinese phone-- They took a big step back two years ago. They didn't even go to advanced. The question is, what does that say about the next 18 months in China? Should we expect to see a big acceleration, both generically and in your revenue, for the mid-tier Chinese phone? What's the competitive dynamic there? Avago tends to avoid that business.
Are they going to come in now because the growth at the top end where they're big is slowing, and this is where everything's gonna start showing up?
Yeah.
Cees, IoT is the buzzword for the day. Everybody's talking about it. It's kind of the Tower of Babel, really. You pointed out a bunch of different technologies. You said yourself it was confusing. The curious thing about it is they're not all gonna win. You talked about GaN MIMO and all that. Most of the application, if you talk to Verizon, AT&T, is in fixed broadband and trying to address or IoT. You've got a cellular technology trying to push its way into what I would call a portable market. At the same time, Wi-Fi and everybody else is expanding and going faster. Doesn't somebody have to lose here? If we're talking about your business, GaN's going up, and you talk about multi-node Wi-Fi, they can't both win, right?
Can you just give us some perspective on who you think's leading and where is that gonna go in the big picture? Thanks.
All right. Starting with China handset dynamics. As you pointed out, China, in fact, is not planning to adopt a non-standalone version. They're going straight to the new radio standalone standard, that will necessarily put them a year later at kind of bringing 5G to market. I think that was your key point, right? When they do, it's liable to come pretty darn quick, when they do bring that to the domestic market. In the meantime, the growth in China is due to the move up higher into the LTE-Advanced capability. Also, we are seeing a lot of unit growth in export markets. Content there can vary dramatically as to which market they're shipping into. But we are seeing the China customer set more and more exporting to various parts of the world, which will continue 4G type content growth between now and 5G.
Just trying to be sure that I understand the question, in general, in the Wi-Fi markets, we are very much on a front-end and filter and GaN focused. GaN is more for the infrastructure market, for the outdoor infrastructure market. Was that your question or?
To the extent that you do MIMO GaN for the infrastructure market, a lot of their targets, Verizon, AT&T, are trying to do fixed broadband applications or IoTs in factories. They're taking that product, and they're saying, "Hey, it isn't gonna work for cellular. Let's try to attack the market." The Wi-Fi's crushed us in year after year, right? They're gonna go out. Whether it's successful or not is a whole another question. People are writing checks to you for GaN MIMO to try to take on Wi-Fi in the IoT market, and Wi-Fi going to AX now, blazing the trail, they both can't win, right?
I think MIMO has multiple applications, though, right? I think MIMO is also just a significantly more efficient base station antenna. I don't want you to just think MIMO is just in the IoT. MIMO is, because of the gain and the number of elements, it's a more efficient way to transmit power for a LTE base station or certainly for a 5G base station. I wouldn't put it all in that category yet. I'd give it both sides. The way I like to think about it is, I had a customer tell me this story that if you think about supplying data to the home and then distributing data inside the home, and there's sort of a going up and down between those two.
We work really hard on the infrastructure, we get all the data to the home, then we can't distribute it. Things like distributed Wi-Fi and AC come along, all of a sudden, now we can move it around the house, now the infrastructure has to go up. This constant up and down, I think, is going to continue across the market and play out in the home. You saw that in Cees's presentation. The infrastructure will significantly upgrade, then we'll play in the house. The great thing about the strategy inside Qorvo is that we play really across all those connections. It doesn't matter. No matter how the infrastructure gets better for me, base station, optical, cable, we're there.
When it gets into the house, as we start going to, whether it be WiGig or AX, we're going to be there to play as well. I think it's a very good balanced portfolio.
Final question, I promise. This is for Todd. Todd, you're going to get out of here with one more at least. I don't know, five, six, seven years ago, we were here talking about 3G, mostly an amplifier story. They had diplexers or duplexers because it was FD. We started moving to 4G, all those bands showed up, duplexers became the theme. Everybody knows that. We talked about bands, all that. Now everybody's kind of still talking about, there's a lot of confusion about, oh, duplexers, as several people have said, the competitors talked about TDD. We're going to 5G, it's got TDD bands. How does the conversation shift? Filters, as you guys pointed out, several presentations, there's a ton of them, there's a whole bunch of new stuff coming up.
Is the story shifting from the duplexer story to, I would say, coexist because we're adding a bunch more bands of 5G. They may not be FD bands, but they're too close to everything else. Are we moving into a coexist environment now and a antennaplexer , which is basically a coexist for the antenna? More importantly, is the whole amplifier story about to change? Because 100 megahertz bandwidth is causing all sorts of problems for you guys. What theme are we going to see more and more of as you move into 5G and the shift between 4G? Thanks.
That's a broad question, as usual. No, 5G is going to be TDD, but it's going to have to, like you mentioned, it's going to have to coexist. If you look at the even 4G bands in China, they're all TDD bands, but they all require a lot of filtering, and that's been around for years, the 2.2 to 2.5 gigahertz in that area. All those bands are TDD bands, but it requires filtering to coexist. 5G, we expect to be something similar from the three and a half to four and a half, even up to the Wi-Fi area, that we expect to have to have filters for coexistence. Yes, that's the theme for 5G. Also 5G is getting retrofitted back to all the FDD bands.
Even those are going to have to coexist because FDD is still going to be in those bands. There's still going to be lots of filtering required. We don't see that changing as they retrofit those bands because they're going to have to coexist FDD with TDD in the same bands. We don't see much of a big change in the filter requirements coming than what they are doing today. Power amplifiers, 100 megahertz, they're tough. To be able to get a power amplifier that can run 100 megahertz bandwidth and be linear, you think of Wi-Fi. If we start doing Wi-Fi type efficiencies in a handset and be able to have to put out a watt or 2 watts of output power, you're not going to want to use your phone for very long. That's something we have to solve.
You're going to be running in the single-digit type of efficiency at 100 megahertz linearity. It's going to look like a Wi-Fi PA with an APT type of power management system. That's why we have to crack this envelope tracking and be able to do envelope tracking at 100 megahertz bandwidth with a PA that has to be able to do 100 megahertz. It is going to be very challenging, it's something that the industry needs to solve.
Okay, seeing no more questions, on behalf of the entire Qorvo organization, we certainly appreciate all of you in the room and those online for attending and your interest in Qorvo, we look forward to updating you throughout the year at future investor events. Thank you, have a great day.