RF tiptoes into the embedded world

DoD has challenged microwave designers with creating subsystems with performance at or in some cases beyond the state of the art - and were just happy to get something that works, even at high cost.

"There are no rules here, we are trying to accomplish something."

- Thomas Edison

This quote from the famed inventor succinctly stated one of the two reasons why microwave subsystems - integrated microwave assemblies (IMAs) - employed in defense systems have never conformed to open standards such as OpenVPX. The second reason is that the U.S. Department of Defense (DoD) never required such conformation, until recently. Designers of board-level embedded products that adhere to specific, fixed form factors and electrical specifications might consider such a scenario chaotic, even ridiculous. To provide some perspective about why it's taken 70 years for the defense industry to tiptoe into open standards, we need to look at both the technology and the DoD.

This quote from the famed inventor succinctly stated one of the two reasons why microwave subsystems – integrated microwave assemblies (IMAs) – employed in defense systems have never conformed to open standards such as OpenVPX. The second reason is that the U.S. Department of Defense (DoD) never required such conformation, until recently. Designers of board-level embedded products that adhere to specific, fixed form factors and electrical specifications might consider such a scenario chaotic, even ridiculous. To provide some perspective about why it’s taken 70 years for the defense industry to tiptoe into open standards, we need to look at both the technology and the DoD.

As engineers designing defense embedded systems know, digital and microwave technology have little in common. Digital systems are designed around highly integrated semiconductor-based components such as central processing units (CPUs), field-programmable gate arrays (FPGAs), general processing units (GPUs), digital signal processors (DSPs), and systems-on-chip (SoCs). The key tasks for the designer are to – within the confines proscribed by a standard form factor – make all these devices work together, program them, deliver as much input/output as possible, and deal with the issues of thermal management and power consumption.

In contrast, while microwave design engineers face some of the aforementioned challenges, they additionally face those unique to this microwave technology as well, the latter dictated by the fundamental nature of “fields and waves.” That is, as wavelength (and thus component size) decreases as frequency increases, subsystems that operate at 500 MHz, for example, are inherently larger than those operating at higher frequencies. Therefore, designing a UHF system to fit in a small, invariable form factor is a challenge that cannot easily be achieved. Digital devices do not suffer from this problem: While FPGAs may get larger with increases in performance, they do not get three times larger. Although microwave subsystems for defense applications typically operate between 1 and 26 GHz where component size is reduced, some components do not get that much smaller.

Microwave engineers also cannot draw from a set of highly integrated devices, like FPGAs, that can perform many functions in a small space. The closest this technology gets to such integration is the monolithic microwave integrated circuit (MMIC) that combines gallium arsenide, gallium nitride, or silicon germanium transistors with tiny components to produce amplification, mixing, and switching, for example. IMAs also must use discrete passive and active components on the circuit board, all of which are connected via etched transmission lines such as microstrip or coplanar waveguide, the length and other dimensions of which are critical. The higher the frequency, the more difficult it becomes to combine all of these circuit elements while maintaining precise matching to 50 ohms, very low insertion loss, and a variety of other essential performance parameters. In short, designing microwave subsystems that must always conform to some declared form factor is exceptionally difficult, and subsystems designed to transmit substantial amounts of power make integrating them into standard form factors like Open VPX even more challenging, and at higher power levels practically impossible.

That said, it’s reasonable to ask how, then, do smartphones transmit and receive data from 700 MHz to 2.6 GHz over the carrier’s network, Wi-Fi, and near field communications (NFC), and receive GPS and possibly FM broadcast signals – all in a package a quarter-inch thick, four inches wide, and five inches high? First, smartphones operate at extremely low power levels and take advantage of SoCs that combine virtually every baseband, DSP, control, and many other functions in a single device. Secondly, the DoD system requirements far exceed those of the commercial market in terms of ruggedness, operating temperature, and an array of other requirements not required by the consumer market. Finally, unlike defense systems, smartphones and tablets must conform to a specified form factor dictated by the end user. In other words, someone (in this case, the user) required it.

In comparison, for the entire history of the microwave industry, the DoD has challenged microwave designers with creating subsystems with performance at or in some cases beyond the state of the art – and were just happy to get something that works, even at high cost. Requiring the microwave subsystem to conform to a 3U or 6U form factor was not even on the table. In other words (as previously stated), “There are no rules here, we are attempting to accomplish something.” As a result, almost every microwave subsystem is unique, designed for a specific system and platform, not the least bit modular, created by the team of designers within a specific company often using proprietary techniques, and extremely difficult to modify once designed and fabricated.

Now, after 70 years of stasis, the DoD is not simply suggesting but in fact mandating that future defense systems be designed according to Modular Open System Architecture (MOSA) principles, with specific attention being paid to “… modular designs with loose coupling and high cohesion.” This mandate is also being reinforced under the latest Better Buying Power 3.0 Implementation Guidance Memorandum issued by Frank Kendall, undersecretary of defense, in April 2015. Under the DoD’s mandate, the Army, Navy, and Air Force each are working to achieve this goal, not only for embedded digital systems, but also for radio frequency (RF) systems.

Mercury Systems is a pioneer in this effort, as it introduced the OpenRFM architecture initiative for RF subsystems in October 2014. It remains – more than a year later – the only such proposal from either the embedded systems or microwave technology domains focused on standardizing various aspects of RF and microwave subsystems. In proposing OpenRFM, Mercury’s goal was not to be the sole manufacturer of such systems but rather to provide a roadmap for others as well, in order to make OpenRFM (or a modified version of it) a recognized standard that supports the DoD’s MOSA principles. The path toward realization of this goal is paved with challenges but the end result will be well worth the effort.

Lorne Graves is a Technical Director with Mercury Systems.