mmWave Modules Address Security and More
Mr. Hebeisen’s responsibilities at Endwave include strategic marketing, business development, investor relations, and implementation of Endwave’s merger and acquisition strategy. He joined Endwave in connection with the September 2002 acquisition of the wireless assets of Signal Technology's Wireless Group (SWG). From 1999 to 2002, Mr. Hebeisen was V.P. of Technology and Business Development for SWG, where he was responsible for engineering and sales divisions of the fixed wireless commercial business unit. Prior to 1999, Mr. Hebeisen was Director of Engineering and Sales, as well as an original founder of Advanced Frequency Products (AFP), a private millimeter wave transceiver start-up company that was later sold to Signal Technology. From 1990 to 1997, Mr. Hebeisen held various RF and microwave design positions for both defense and commercial wireless applications with Alpha Industries and Raytheon. Mr. Hebeisen holds both BSEE and MSEE degrees from the University of Massachusetts at Amherst, and is a graduate from the AeA/Stanford Executive Institute for Management of High Tech Companies.
MWJ: Endwave designs and manufactures customized microwave components and subsystems up to 100 GHz using hybrid MIC circuits, commercial and custom MMICs, as well as proprietary flip-chip technology. Your products are used in high-speed cellular backhaul networks, enterprise access, commercial radar systems as well as homeland security, government, and other defense electronics applications. Starting with your commercial telecom business, can you tell us - what are some of the big market drivers for microwave point-to-point radio these days?
MH: For Endwave’s telecom business, our high-frequency transceivers typically end up in cellular “backhaul” networks that connect base stations to the core network or switching centers. In this market, growth is based on two underlying, but separate, drivers of equipment deployments. First, there are 2G voice-centric network installations in developing countries where telephone network infrastructure simply does not exist, or is of insufficient quality. Many of these regions are in the midst of massive network build-outs as a result of skyrocketing growth in mobile subscribers -- for example, India added over 50 million new subscribers to cellular service in just the first half of this year! But going on in parallel is an evolution towards data-centric network enhancements in developed countries. These network enhancements are required to meet the transmission capacity and capability demands of HSDPA, WiMax, LTE and other bandwidth-intensive data transmission standards. As the deployment of broadband applications accelerates and backhaul traffic increases, the demand for higher data-rate microwave radios at key aggregation points is on the rise. In turn, as a leading supplier of microwave transceivers that form the “heart” of these radios, wireless OEMs turn to us to help them address this insatiable thirst for network bandwidth.
MWJ: In which cases do a microwave link win over a wire-line transmission network, are these networks primarily targeting regions where an existing wire-line infrastructure is non-existent, say some parts of Africa or South America or is the need being driven by new performance requirements such as capacity or operating costs?
MH: It’s interesting to look at different parts of the world and see the battle between wireless and wireline technologies. In the North America, the backhaul has historically been provided by the local wireline carriers in the form of leased T1 line (ie, copper that allows ~ 1.5 Mbps transmission). So only 10-20% of the base stations today are using microwave radio backhaul in this part of the world. However, one finds that as the bandwidth requirement at any particular base station increases above several T1 lines, the economics of wireless backhaul begins to shine. Microwave radios provide carrier-class quality of service and can be installed in a matter of hours – try getting that type of speed with the wireline carriers on T1 installations, where the wait can easily be measured in months. In Europe where they tend to be ahead of the curve on mobile penetration and advanced wireless services, we believe that over 80% of the base stations are using microwave backhaul to connect up their cell sites to the switching network. And in emerging markets like Asia, South America and Africa, there are currently large build-outs of microwave radio backhaul networks given their fast install, high-reliability, lower cost, and elimination of trenching costs associated with wireline technologies.
MWJ: Are Endwave’s telecommunications products also targeting backhaul for 4G networks?
MH: Yes, they are. We expect an increasing amount of microwave radio backhaul as data-intensive applications like WiMax and the proliferation of advanced bandwidth-thirsty wireless devices like the iPhone put incredible strains on legacy backhaul networks. Leased-line T1s (and E1s elsewhere, ~ 2 Mbps) inherently provide for a circuit-switched infrastructure that is not optimized for the demands of next-generation 4G services that are packet-switched, by nature. In today’s wireless network deployments, backhaul solutions are based on what is referred to as TDM technology, including low-to-medium capacity PDH radios and higher capacity SDH radios. There is a large installed base of these products, and most of the million or so point-to-point radios deployed last year, for example, fell into this technology group. However, as wireless data usage escalates and accounts for an increasingly larger portion of overall network traffic, wireless carriers are turning their sights to Ethernet–based solutions, or IP radios. Internet Protocol technology can yield more scalable solutions that also enable a more efficient means of transmission. Endwave is well-positioned with wireless OEMs in assisting their development of hybrid Ethernet/TDM microwave radio products that offer carriers a seamless migration path from TDM to IP-based backhaul.
MWJ: There seems to be an uptick in the people’s interest in millimeter-wave technology such as “E-band”. If so what’s driving this renewed interest and is it weighted more toward commercial, military or government end-use?
MH: Millimeter-wave technologies are increasingly called upon for advanced defense electronics, homeland security and telecommunications applications. This is due in large part to their unique ability to carry large amounts of data, to provide higher resolution for radar systems, to “see through” certain structures, and to provide solutions that are small, light-weight, and high performance. For the emerging E-Band market, new radios at 70/80 GHz are being used in a wide range of applications. On the commercial side, Endwave participated in the 1st commercial deployment of an E-Band gigabit point-to-point radio link to provide quadruple-play services to residential customers for high-speed data, voice/video over IP, high-definition TV, and video-on-demand service. WiMax backhaul is also and area that may have significant impact on E-Band radio volumes. But some also look to E-Band solutions for disaster recovery communications, and providing a broad data pipe between government buildings.
MWJ: As we mentioned, Endwave operates two complimentary business units - a commercial telecom business unit and a Defense/Security unit. Can you explain more how these two distinct but important businesses actually complement one another?
MH: Believe it or not, many of our closest customers in the defense electronics business were originally drawn to Endwave for our successes in commercial microwave and millimeter-wave markets. We’ve been a leader in the telecom space for a long time, with over half a million high-frequency modules delivered to such applications. There aren’t many companies who can tout such an achievement. Defense OEMs have made a conscious effort to harness that experience and innovation that Endwave has brought to the telecom market – by asking us to apply those advancements specifically for stringent military and government applications. The continuous exchange of technology breakthroughs and manufacturing process improvements that flow between our defense and commercial business units solidifies our leadership position in both arenas. Our MLMS flip-chip technology, as an example, was originally developed under a defense R&D project – and now is a critical technology for our 70/80 GHz commercial E-Band transceivers. These complementary activities bring healthy balance and stability to Endwave, helping us to become a more reliable high-frequency partner for a broad range of customers.
MWJ: I assume each of these groups have specific challenges to developing the leading edge innovations required by today’s defense and commercial applications, such as time to market or ruggedness? Do these groups have specific strengths based on the customers and markets they serve?
MH: Mission critical defense and security applications certainly call for impeccable performance and quality levels, as failure is not an option. However, we take that same mentality on the commercial products as our large OEM customers inherently demand top quality, as they should. Certainly millimeter-wave proficiency is a core competence that is inherent to both our business units, so we tend to focus on using that strength to provide solutions to our customer’s most challenging problems. While the specific challenges our two groups face can vary, our same commitment to bringing innovative technology, manufacturing efficiency and quality is what separates us from other companies. So, there is more commonality here than some might think.
MWJ: The Defense group was formed in 2005, was there some aspect of delivering products for defense and homeland security that made it easier to address with a separate business unit from your commercial side?
MH: First, we needed to assure our defense customers that we were going to be attentive to their every need, just like we are for the likes of large OEMs like Nokia Siemens Networks on the telecom side. Our mission back then was to concentrate and consolidate the technology and manufacturing leadership from both JCA Technology (newly acquired at that time) and Endwave’s emerging government end-use business under Endwave Defense Systems. By leveraging this world-class portfolio of products into highly integrated solutions for defense prime contractors, we were well-positioned to garner an increasing share of the defense markets while bringing added balance and stability to Endwave. And we’ve done just that. Last quarter, we reported over $5 M in non-telecom revenues – some 30% of our overall quarterly revenues. Those revenues are to customers and markets that we were not really serving five years ago, so we are proud to have created a substantial business base here.
MWJ: I read your engineering team has developed an extensive IC and module design library which they reference to provide rapid design prototypes of new RF subsystems for your customers. Could you tell us a little more about the design process? For instance – how does your engineering team use this existing design library to create a new multi-functional assemblies (MFAs) based on customer specifications.
MH: Surely. At Endwave, the typical product we provide is a microwave or millimeter-wave module that enables higher level radio or radar architectures. However, behind that subsystem prowess is an internal MMIC design capability that uses commercial foundries around the world to produce world-class semiconductor chips that make our subsystems better. So we literally cover the high-frequency industry from chips to systems – we call it our systems mentality with a semiconductor “soul”.
As such, we have accumulated an extensive design library of circuits and full-up modules that allows us to quickly and efficiently customize multi-function module solutions. But sometimes even a large design library won’t do the trick. One example of how our internal semiconductor design capability was leveraged to create something evolutionary was our development of a Ka-band PIN diode SPDT switch MMIC. This custom chip design was undertaken to support the development effort for a cost-effective, high-power, extremely compact, Ka-band multi-channel radar transceiver module. The up-front portion of the design process focused on the evaluation and, ultimately, the exclusion, of other available switch technology and topology options. Subsequently, when no previous Endwave building-block nor commercially available solution was available to do the job right, we embarked on a bold mission to custom-design a high-power, fast-switching Tx/Rx MMIC switch. In a matter of months, we created a chip that provides well over 4 Watts of power handling capability, exhibits nanosecond switching speed, and paved the way for a condensed T/R module footprint -- making it ideal for emerging Ka-band seeker applications. While there are other large companies out there that might boast such vertical integration, few can pull that off and still behave like a nimble, fast-paced innovator that is a reputation Endwave has earned over many years.
MWJ: Do most of your designs evolve from existing ones?
MH: In many cases, yes. When you consider that we’ve got several thousand different RF module variants in our library plus a wide array of circuit building blocks, it’s pretty rare that we don’t have an effective starting-point. We know how critical time-to-market is to our customers, so they turn to us knowing that we aren’t going to have to re-invent the wheel every time they want a module customized to their exact needs. Our amplifier business is a perfect example of this. We have three strong amplifier product lines within Endwave, thanks to key acquisitions that include TRW/Milliwave, JCA Technology, and ALC Microwave. We sell plenty of catalog amplifiers from those established product lines. But it’s also very common for us to hear from customers – “I want that exact model number….but can you put a limiter at the front, or can you add a power amplifier stage at the output?” So it is still very much a customized type of business – and we’ve geared our engineering, manufacturing and customer service groups to play to that fast-paced market need.
MWJ: You must work very closely with your customers. Is there a lot of back and forth between your engineering team and the customer? Do your engineers help them define subsystem specifications and determine how to implement the appropriate technology or are most of your customers looking for a black box that delivers specific performance?
MH: It’s very much the former – we have very heavy collaboration with our customer’s engineering team in creating the specifications for the product at hand. And we even steer them in new directions for their system architecture based on our knowledge of the challenges they face. We truly understand how our units play at the higher system level, giving us the ability to recommend cost vs. performance advantages in their system architectures that impact modulation options, specification levels, frequency planning, and mechanical layout. Then, to be able to address those system challenges down to the chip level – that is the value we bring to the table. That complete understanding from the top-level system architectures down to the device physics – our customers tell us how critical that perspective is to them on front-end collaboration.
MWJ: Are most of your designs driven by customer request or do you develop some “generic” components and subassemblies for application in markets that look promising?
MH: A little of both, as it varies based on the markets that we serve. For our amplifier business, for example, it’s quite typical that we produce a catalog design that we believe will appeal to a wide range of customers and markets. So you might call those “generic”, to use your word. But the technology that goes into those products is anything but “generic”. On the flip side, for our core telecom transceiver business, we provide the integrated Tx/Rx module that forms anywhere from 80-90% of the electronics that go into an Outdoor Unit, or ODU, that goes up on the pole above a base station. Given the wide range of diplexer options and configurations that our customers use to interface with the transceivers on the high-frequency RF side, and the wide range of baseband processing and IF circuitry our customers use on the back-end of our modules – “standard” modules are rare in this business. One exception to that is our new E-Band Tx/Rx pair that we designed for this emerging multi-gigabit data communications market. We’ve sold that E-Band T/R module to many different radio providers as a “standard” product. So you can see, it’s not a “one-size-fits-all” business.
MWJ: Developing one-of-a-kind prototypes for new systems must have your engineers exposed to a lot of next generation telecommunication and defense equipment. Is there considerable latency between your initial product development and the deployment of these systems?
MH: On the defense side – yes, there can certainly be some long gestation cycles between design and full-blown production. On the telecom side – not so much. In defense electronics, there is often a long development / qualification period for many of our complex subsystem platforms that can last 1-2 years. Production might not happen until years in the future. So you’ve got to pick your opportunities wisely, and execute. We recently announced an 18-month multi-million dollar production contract from Boeing to supply broadband frequency converters used in the modernization of the US Air Force’s AWACs 3035 system. AWACs provides survivable airborne surveillance, command, control and communications functions, plus early warning detection and tracking of low-level targets at extended ranges over land and water. The 3035 program is a modernization upgrade to AWACs, and our units go into the Electronic Support Measures, or ESM, portion of the upgrade. These products were designed and qualified several years ago – yet here we are now getting our first production contract. Deliveries have already commenced, and this is a great example of a defense platform that we began investing in years ago and are now enjoying volume production. In telecom, the entire lifetime of a microwave radio family can often be measured in 4 or 5 years. Furthermore, there can be over a dozen different frequency variants to fully cover the necessary frequency spectrum used to blanket the world. So in this space, it’s critical that you deliver on-time, the first time, or you risk losing a market window.
MWJ: You make products for atmosphere profiling, airport traffic control and severe weather forecasting (referring to k and V- band down converters released last year). Do these products fall under the defense group’s responsibilities?
MH: Yes, but more so because of the manufacturing resources that are necessary to produce these critical products. Our defense products are all manufactured domestically in a factory outfitted with multiple lines of automated assembly equipment (Palomar). As the frequencies creep higher and higher, the use of automation to build millimeter-wave circuits becomes increasingly important. We use this Palomar automatic pick-and-place machinery to build the K-Band and V-Band products with precision and repeatability. Even minor variations in wirebond lengths and gaps between components can wreak havoc on products operating at millimeter-wave frequencies, so manual assembly techniques simply fall short in this realm. Endwave’s automation allows us to build unit number 500 with virtually identical performance to unit number 1.
MWJ: One interesting millimeter-wave application is “whole body imaging” that is being developed by L3 Communications for security and detection systems useed in places such as airports and government facilities. What’s Endwave’s contribution to this system and will they eventually replace metal detectors?
MH: I think our relationship with SafeView (part of L-3 since they were acquired in 2006) is one that deserves some special attention. We initially began our relationship with this private Santa Clara start-up by designing the Tx/Rx modules that went into their first generation scanner. This was straightforward for us, given our leadership in telecom transceivers and the similar technology and manufacturing processes used in our core business. Soon after, we were awarded the switch modules that allow the transmit signal to quickly switch between antenna elements and scan the body from floor to ceiling. Still later, SafeView awarded us the entire RF “mast” – or chassis, of which there are two inside any ProVision™ portal. These 7-foot tall chassis include all of our electronics, plus all the associated cabling, connectors, and metalwork. We integrate all that, run final test, and deliver these large masts to L-3 SafeView where they do the final integration into the portal and ship it out the door. So you can see we are far more than your everyday module supplier.
In April of this year, the Transportation Security Administration Chief, Kip Hawley, publicly announced that the TSA was going to ramp up purchases of millimeter-wave checkpoint security equipment. Specifically, the L-3 SafeView ProVision™ portals were singled out for deployment in many US airports. During Q2 of this year, the TSA began deploying 38 ProVision™ scanners in LA, Baltimore, Denver, Albuquerque, Dallas, Detroit, Las Vegan, Miami, and New York’s JFK airport. Even more encouraging were comments from other TSA officials stating that the scanners could eventually replace metal detectors at the nation's 2,000 airport checkpoints and the pat-downs done on passengers who need extra screening. In addition to these airport checkpoint deployments, L-3 has publicly stated that scanners are already used in many different locations, including a few courthouses, jails and US embassies, as well as overseas border crossings, military checkpoints, government buildings and some foreign airports such as Amsterdam's Schiphol. So, it will be difficult for anyone reading this article to NOT see these millimeter-wave portals if they do airline travel. That should be exciting to our entire industry, as it’s sort of like “millimeter-waves going mainstream”.
MWJ: That is very interesting and I look forward to seeing this technology showing up in airports, especially if it means getting through security faster. Speaking of millimeter-waves going mainstream, this month’s Microwave Journal features your new 60 GHz T/R module that targets applications that require extremely high data rates such as wireless streaming high definition television. This product got a lot of press and resulting interest. Is this the work of your commercial group? Can you share some insights on this application and market potential?
MH: Yes, our commercial group ran this project and the interest we’ve generated with some pretty large “players” has been fantastic. Even more exciting is the large number of applications we are hearing from potential customers what our 60 GHz T/R could be used for – while it would be premature for me to talk about these ideas or the customers, suffice it to say that there seem to be boundless opportunities to wirelessly route multi-gigabit data streams in an indoor, short-range environment. The application that has drummed up a lot of interest is to replace the expensive HDMI cables that we all have connecting our High-Definition set-top boxes to our HDTVs – and the interconnections between multiple multi-media devices in the same home. So picture wirelessly connecting such devices, eliminating the rats-nest of cabling currently going to our consumer electronics – and doing so without sacrificing picture or sound quality? 60 GHz can do that.
MWJ: Do you have engineering resources focused solely on advanced module integration, packaging and interconnect technology such as Endwave’s proprietary technologies, MLMS™ and Epsilon™ Packaging? Can you tell us a little about these innovations?
MH: Yes, we have an advanced R&D group that works independently from our product design groups. But there is regular communication between groups so that we can adjust to market and technology trends that evolve all the time. As our new commercial technologies are ready to go “live” into actual products, we have an efficient hand-off process that allows for an easy implementation of advanced technology. You highlighted a couple of our key technologies, so let me elaborate there.
Conventional millimeter-wave thinking is rooted in the belief that module housings have to be fabricated utilizing costly machining and plating techniques. Endwave’s Epsilon™ Packaging eliminates this dependency by replacing costly and heavy weight metal mechanicals with metallized FR-4 and injection molded metallized plastics. Epsilon includes revolutionary mixed technology integration, allowing chip-on-board and surface-mount technology components to co-exist, thus easing assembly and reducing cost. The end result is a package with no machined metal parts and is mass producible with efficient heat extraction and minimal weight and size.
Turning to Multilithic Microsystems™, or MLMS, this is a flip-chip semiconductor technology that is particularly useful in high millimeter-wave frequencies. A lot of folks claim flip-chip expertise in the lower RF frequencies – but few can pull it off at frequencies over 100 GHz. This advanced circuit technology uses flip-chip and electromagnetic coupling methods to minimize expensive semiconductor real estate while eliminating wirebonds and their related variability. MLMS moves passive circuitry onto an inexpensive proprietary substrate that processes with the ease of silicon, yet works past 100 GHz. Only the discrete active FET or PHEMT devices remain, which are then flip-mounted on top of the MLMS substrate “bumps”. With MLMS, a complex transmitter or receiver can be placed on a single, compact substrate with no bondwires in the RF path. Our E-Band Tx/Rx pair uses MLMS to create what we believe are the best performing 70/80 GHz mixers in the industry.
MWJ: Sounds like a great blending of different technologies to achieve high-performance and reliable manufacturing, which must help with costs. Speaking of which, Endwave has an impressive set-up for moving engineering prototype development to small volume production or high-volume manufacturing. Is it difficult to transfer technology between manufacturing facilities?
MH: Excellent question. We proudly offer both domestic and offshore manufacturing of high-frequency components and subsystems. Most products manufactured in our US operations are intended for defense, homeland security, and other government end-use applications. Our secure facilities are ISO9001-2000, AS9100 and SCIF certified. Since the year 2000, the majority of our commercial product lines are produced at our contract manufacturing partner in Thailand. A talented Endwave support team is ever-present to provide constant oversight. Our team was the first to outsource high-volume millimeter-wave production from the US to Asia, so the industry recognizes us as a true pioneer in offshore manufacturing -- creating a scalable manufacturing “engine” for low-mix, high-volume applications. For non-DoD products, there are always parts transitioning from our domestic factories to our offshore team – so we use stringent design for manufacturability (DFM) techniques to ensure a smooth transition. Many of our modules can be manufactured either domestically or offshore, giving us flexibility and disaster relief plans that provides our customers additional comfort.
MWJ: I’ve read about Endwave’s use of RFID tracking in its production line to manage component selection and assembly. Does this reflect the complexity of manufacturing your leading edge modules?
MH: We were extremely proud to have received not one – but two awards for manufacturing innovation for our implementation of automated RFID sensors to track our work-in-progress (WIP) subassemblies on our manufacturing floor in Diamond Springs, CA. We see the implementation of cutting-edge RFID tracking systems on our factory floor as another example of Endwave’s desire to apply automation in clever ways to ultimately improve our efficiency and customer service. Specifically, Endwave received the “Progressive Manufacturing Award” by Managing Automation magazine, followed just weeks later by the Gold medal from M2M, or “Machine-to-Machine”, magazine for the “M2M Value Chain Award”. This really underscores that Endwave continues to look for ways to provide our customers a level of service and security they probably won’t get elsewhere.
MWJ: Endwave was originally incorporated in California in 1991 as Endgate Corporation. The company changed its name to Endwave Corporation after merging with TRW Milliwave Incorporated in 2000. After going public in October of 2000, Endwave made numerous acquisitions including the Stellex Broadband business from M/A-COM, Inc.; the Fixed Wireless Division of Signal Technology Corporation, a product line from the broadband business of Arcom Wireless, Inc., defense synthesizer (MTS-2000) assets from Verticom, Inc., amplifier products through the acquisition of JCA Technology from Bookham Technology. In 2007, you added ALC Microwave to your family. These mergers have made Endwave one of the largest commercial suppliers of microwave and millimeter-wave RF subsystems. I noticed part of your responsibilities as VP of business development is to implement Endwave’s merger and acquisition strategy. What were you looking for in the businesses you have acquired?
MH: I’ve been involved in M&A in varying degrees for the past 10 years, from the time I led the sale our private start-up, AFP, to Signal Technology. At Signal, I worked on their M&A team, and led the sale of the wireless group to Endwave back in 2002. I was proud to take over such M&A efforts at Endwave back in 2006. Our last acquisition of ALC Microwave in April of 2007 has been very successful thus far. We saw in ALC a great product company with a niche in the defense business, but one whose growth could be substantially accelerated by wrapping the technology, manufacturing and marketing strengths of Endwave around them. We are very proud of ALC’s contributions to Endwave thus far. In addition to the log video amplifiers which ALC has been known for, they have also brought Endwave new contract wins with several customers for driver and power amplifier products. So we are pleased to see that ALC is providing us new products, customers in both domestic and international defense markets, and opportunities that extend well past their core log amplifier business. Going forward, we continue to look for companies in the microwave / mm-wave space who provide similar opportunities for accelerated growth in all the markets we serve.
MWJ: Is there a certain technology not in your current portfolio that you are keeping your eye on to see how it will evolve?
MH: Our MMIC design capability continues to improve our ability to provide world-class subsystems to our customers. Given that MMIC experience, we have relationships with most of the commercial semiconductor foundries out there – from wafer producers of GaAs, InP, SiGe, you name it. Personally, I am keeping my eye on the Silicon (CMOS) progression into the millimeter-wave space. There are several applications emerging where substantially less-expensive semiconductor technologies may provide a level of performance that is “good enough”. The up-front costs of using such technologies are often quite expensive, so one must be careful heading in this direction. For this reason, I believe Endwave will continue to focus on the III-V semiconductor technologies while keeping an eye to the future. But it will not surprise me to see SiGe and Si begin to find a place in the microwaves and millimeter-waves that has traditionally been a “GaAs” game.
MWJ: Well Mark, I certainly appreciate your taking time to share your thoughts with our readers. Endwave is clearly at the forefront of technology and one of the hardware providers behind the news stories such as the explosive growth in point-to-point radios for cellular backhaul, Wireless HD or new homeland security products that our readers might have the chance to personally encounter in the field. I wish you the best of luck. Thank you.
MH: Thank you, David