GaN adoption, RF & microwave funding growing in military market

Every month the McHale Report will host an online roundtable with experts from the defense electronics industry – from major prime contractors to defense component suppliers. Each roundtable will explore topics important to the military embedded electronics market. This month we discuss radio frequency (RF) & technology trends, (GaN) adoption within military applications such as , defense funding for RF and microwave technology, and the buzz on the floor of the International Microwave Symposium (IMS) in Honolulu, Hawaii last month.

This month’s panelists are: Gavin Smith, RF Industrial Product Marketing - Business Development Manager WW at NXP Semiconductors; Ryan Baker, Product Marketing Manager, RF Components at Wolfspeed, a Cree Company; Mike Ziehl, VP and General Manager, Multi-Market Business Unit, MACOM; and Philip Fulmer, Director, Product Marketing at Mercury Systems Advanced Microelectronic Solutions group.

MCHALE REPORT: The International Microwave Symposium was held last month in Honolulu, Hawaii. What was the buzz on RF and microwave technology at the event?

SMITH: Similar to 2016, 5G was a major topic with many exhibitors showing advancements and improvements from the year before. 5G is so fresh that many companies are cognizant and secretive about what they put on display at these types of events. The 5G keynotes were interesting with Professor Henning Schulzrinne from Columbia University and Dr. Wen Tong from Huawei Technologies. Professor Schulzrinne provided his insights on the future of data and the role of public policy. In his view, generations from baby boomers to generation X to millennials are distinct and talk a different language as they step into using innovation. One of Dr. Wong’s interesting slides stated that “5G is to unleash a bigger bang” showing 2007 applications we use and know well today, but what will 2027 bring?

[Gallium Nitride] GaN technology continued to increase with the number of GaN products on display ranging from cellular base stations to communications and radar end applications. GaN is now a proven technology with established long-term performance and reliability metrics. According to ABI Research’s RF Power Semiconductor Devices for Mobile Wireless Infrastructure report, GaN RF power devices are expected to represent nearly 25 percent of all high-power semiconductors for mobile wireless infrastructure in 2017. 5G will fuel the market’s growth for the next five years and also heat up the GaN market.

The RF Energy Alliance (RFEA), which NXP is a member, had a presence at IMS with a vision for the future using solid-state RF as an energy source for emerging applications. As we come upon the 50th anniversary of the microwave oven, we anticipate solid-state devices will expand the functionality of microwave ovens to be versatile for cooking multiple types of food simultaneously with precise results.

BAKER: Most of the technical program and the product discussions centered around 5G telecom. As an RF and microwave conference, this makes sense due to the significant impact 5G will have on wireless infrastructure and new opportunities opened for communications at 4GHz through mm-wave. Workshops addressed advances in Si CMOS for mmWave 5G, as well as new digital pre-distortion techniques (DPD) for high power amplifiers.

In addition, the growing interest in unmanned vehicles drove discussions on automotive radar advances and next generation sensors for UAVs [unmanned aerial vehicles]. There was a very interesting technical session about radar for maritime applications presented by U.S. Naval Research Lab. It described an X-Band, multi-channel synthetic aperture radar (MSAR) with full MIMO (multi-input, multi-output) capability.

ZIEHL: As expected there was a lot of excitement about the accelerating rate of adoption for GaN-on-Silicon-based RF devices within 4G , and tremendous interest in RF Energy enabling technologies targeted for use in next-generation solid state cooking, lighting and heating/drying applications, among other applications. For the military and aerospace community, hi-rel [high-reliability] components were again front and center in many of our discussions.

One area where we saw a lot of overlapping interest from the military and commercial communities was AESA [active electronically scanned array] radar. Electronically steered active antennas have emerged as a key enabling technology for 5G wireless infrastructure, given the massive MIMO antenna configurations and advanced beamforming capabilities needed to deliver 5G data rates. There are many parallels between AESA radar and massive MIMO 5G systems in terms of architecture and assembly.

Of course, one doesn’t typically associate military radar and commercial basestations where cost structures are concerned, but companies like MACOM have been working hard to close the gap. Using highly-integrated antenna sub-systems and commercial manufacturing techniques, flat-panel Multifunction Phased Array Radar (MPAR) systems have already demonstrated 5X cost reductions compared to legacy slat-array radar systems. Among IMS attendees there was a general consensus that continued innovations in technologies like MPAR will help to enable 5G infrastructure.

FULMER: 5G and products were the buzz at IMS this year, as huge amounts of data traffic are expected to be generated in coming years as the number of connected devices providing real-time sensor data, security data, communications, and military intelligence information continues to increase.

MCHALE REPORT: GaN adaption continues to be the hottest trend in the industry especially within military applications, but there also seems to be more education needed for the customer base on its benefits and where and when to use it? Do you agree and if so how do you go about the education process?

SMITH: First, with any emerging technology there is always going to be some resistance since engineers are comfortable with what they know best. Although GaN and LDMOS have been around for quite some time now, I believe that customers are now seeing the value that GaN can bring to their end applications.

At NXP, our bread and butter is LDMOS technology but we have been using GaN more and more. For our military applications, LDMOS enables high power and good performance up through L-band. As we begin to push to S-band and potentially in the future to C-band and X-band, we see a need to use GaN technology to reach higher frequencies and in some cases higher power for these applications.

Part of the education process at NXP has consisted of GaN workshops, technology comparisons and studies of GaN and LDMOS to understand the benefits and tradeoffs of each technology. These practices help us further understand GaN and gives us the ability to create marketable products.

BAKER: GaN’s properties lend it to multiple semiconductor applications. Blue GaN LED’s took off in the mid-1990s, followed by RF GaN MMICs and transistors in the 2000s. There is talk today for GaN for Power switching applications, but this looks more applicable to the sub-600V Schottky diode market. Thanks to the collaboration between government and industry, GaN for RF applications is in mass production in multiple markets. For the RF Power transistor market in general, GaN-on-SiC HEMTs have about 30 percent market share and are estimated to achieve 50 percent market share in the next five years. GaN HEMTs are taking market share from Si LDMOS, Si HBTs, and GaAs pHEMTs on the semiconductor front. However GaN RF devices will grow the size of the total market for RF Power transistors too; by offering high power solid-state power amplifier (SSPA) solutions with improved performance, over incumbent traveling-wave-tube (TWT) and magnetron based PAs. There will be a place for vacuum electron device (VED) based PAs, but this will erode over time, due to more reliable, higher performance SSPAs.

We continue to advance the GaN RF semiconductor technologies to capitalize on the compound semiconductor’s properties. This means we need to educate industry participants on how to get the most out of our devices. For example, an engineer with experience designing with enhancement mode devices, like silicon LDMOS, needs to take into account GaN HEMT’s negative supply voltage, as they are depletion mode devices. We also educate engineers designing with TWT amplifiers on how to make the jump to more reliable SSPAs.

Educating an industry takes discipline to understand the problems you are solving, show the system level benefits from GaN-on-SiC RF devices, and then provide tools to engineers in order to implement the solution. In the military world, does the application require a smaller jammer, improving SWaP-C (size, weight, power, and cost)? Does a UAV require lighter PAs so it can have a longer flight mission? Does a aircraft benefit from AESA PAs making it more difficult to detect? All these benefits come from implementing GaN RF devices.

RF GaN-on-SiC cannot be beat for bandwidth and efficiency for output power levels >2W. Above 3GHz, GaN-on-SiC competes with Gallium Arsenide (GaAs) pHEMT and TWT technologies. In the beginning, GaN RF Power devices’ higher costs and lower volumes meant that the first adopters were mission critical, broadband applications such as , replacing GaAs and TWTs. Over the last >5 years, GaN-on-SiC RF Power devices have overtaken Si LDMOS in many narrowband application, sub-3GHz, as costs have come down and network operators demand more efficiency and capability from hardware. This narrow band adoption is aided by once novel PA techniques, like Envelop Tracking, Asymmetric Multi-outphasing, and Doherty designs. The aforementioned techniques are more for linear PAs, but even upgrades to legacy narrowband IFF and DME PAs are upgrading to GaN RF Power devices.

ZIEHL: There are several important factors to weigh when deliberating about where and when to use Gallium Nitride (GaN), but the customer learning curve isn’t nearly as steep as it used to be. GaN is a mainstream technology now, and there’s ample industry data available that shows the relative merits of GaN-on-Silicon compared to GaN-on-Silicon Carbide and legacy LDMOS. The performance benefits that GaN-on-Silicon delivers compared to LDMOS are unequivocal, and their cost structures can be comparable at scaled volume production levels for GaN-on-Silicon. GaN-on-Silicon is the clear winner here, and could displace LDMOS in most applications going forward.

In comparisons between GaN-on-Silicon and GaN-on-Silicon Carbide, GaN-on-Silicon is the clear winner in cost structure, manufacturing capacity and supply chain flexibility. Further, industry data shows that GaN-on-Silicon delivers comparable and in many cases superior performance in key attributes like power efficiency, thermal conductivity and reliability.

It’s fair to say that there are a couple of niche applications within the military and aerospace domains where a slight incremental gain in performance may be worth the cost penalty of GaN-on-Silicon Carbide. Generally speaking though, it is anticipated that GaN-on-Silicon could render GaN-on-Silicon Carbide obsolete for high-volume applications.

FULMER: Some early GaN adopters experienced linearity challenges and we still see this perception lingering today. It takes a focused effort to clearly articulate the advances GaN technology has made since the introduction of those early solutions.

As with any technology, the critical factor to a successful implementation lies in understanding the application-specific requirements. There is rarely a universal one-size-fits-all approach and GaN technology is no exception. What works for one application is not a guarantee it will work in another. In some cases, an off-the-shelf existing solution may work. In another case, a custom solution with application-specific specifications may need to be designed.

MCHALE REPORT: Do you see continued growth for RF and microwave technology in military applications with the announced increases in the Department of Defense’s (DoD’s) budget by the Trump Administration?

SMITH: RF is everywhere, when we see an increase in DoD spending, we will see continued growth in RF and microwave technology. I would assume with the increase in spending they will make upgrades to current systems that are using RF technology. This could also open the door for RF to replace outdated technology like vacuum tubes, enabling even more potential growth. Although there is a projected increase in spending, it will take time for programs to take off as it always does for the military market. We are excited to work closely with our customers on new programs and system upgrades and for innovations in research and development.

BAKER: The president’s budget proposes an increase to DHS of about six percent and DoD of about 10 percent. The proposals include the purchase of F-35s, increasing the number of naval ships, and increasing the size of the Army and Marine Corps. As it relates to these specific increases, there will be new RF technology in the communications, radar, and electronic warfare functions of aircraft and ships.

ZIEHL: Today there is activity across the board in the RF and microwave industry, and we expect to see this growth naturally benefit from the announced increases to the Department of Defense’s budget. We, at MACOMx look forward to this continued growth and our role in it. With the existing growth areas and the increased DoD spending, the opportunities for GaN in military applications will grow as well, be it in broadband communications, networked battlefield, radar or electronic warfare.

FULMER: Regardless of the specifics of the defense budget that is approved, we feel confident there will be continued growth for RF and microwave technology within defense applications. Looking at an example, we know that the military will continue transitioning away from conventional bombs towards to smarter ones, minimizing collateral damage and increasing precision strike capability. Guidance, navigation and control systems for smart munitions offer tremendous growth opportunities. Manufacturers with RF/microwave design and manufacturing capability to produce extremely rugged, miniaturized RF/microwave processing solutions for these requirements should be well placed.

We are also confident that investment in electronic warfare will continue. At one point in a time, the battlefield was a physical confrontation between two enemy forces. Today, we are threatened with invisible attacks via electronic warfare. The fundamental nature of the battlefield has evolved, and will continue to further evolve – it simply can’t be ignored. The only way to address the evolution of this changing threat environment is with a substantial investment in RF and microwave technologies to advance our electronic warfare capabilities.

MCHALE REPORT: What military applications are the best bets for RF and microwave suppliers? Radar? Electronic warfare (EW)? ? Other?

SMITH: In the military space, we have success and great interest in communications and radar applications. For RF, communications is such a large target market because it encompasses electronic warfare, radio communications, and [satellite communication]. This is also an intriguing space for NXP RF Power because we can leverage our knowledge from the cellular base station market and apply that expertise to military applications as well. With GaN technology, we see more potential in the EW, radio communications applications that require wider bandwidth and higher power. Aside from communications, L-band and S-band radar are still very important as well. We see continued interest in high power devices for radar applications as older systems start being upgraded.

BAKER: It’s difficult to pick a best, when the answer is “it depends on what you want to achieve.” RF and nicrowave is broad area, so I will limit my response to military applications needing more than a watt of power. Radar, in general, has the broadest market for RF with UHF, L-, S-, C-, and X-Band frequencies all seeing upgrades from legacy technologies which have less performance. It’s the largest RF market within the military arena. New systems will be funded and RF will play into that funding as a driver.

FULMER: Radar, EW, and unmanned vehicle applications each present opportunities for RF/microwave solution manufacturers, with EW applications being particularly prevalent. Our adversaries continue to develop sophisticated agile capabilities within the electromagnetic spectrum domain. To remain in front and continue to dominate the spectrum, RF/microwave manufacturers can’t continue to produce solutions without looking at the broader picture. We will see advances in RF technologies, coupled with tighter RF and digital integration driving EW capabilities. The few RF suppliers that can bridge the gap between the and digital worlds will benefit the most.

Coming back to the ability of our adversaries to quickly deploy agile capabilities threatening our EW leadership position, it’s not enough for RF suppliers to think in terms of single-generation solutions anymore. To outpace the threats of our adversaries while maintaining affordability, the industry needs to adopt a modular, open architecture standard. Mercury’s OpenRFM initiative was designed to address this particular need within the electronic warfare and signal intelligence communities.