Military Embedded Systems

COTS: 10 years after - Well, sure...but what about the next 10 years?


April 01, 2006

Chris A. Ciufo

General Micro Systems, Inc.

(Unofficial) electronic battlefield doctrine I see three key tenets of DoD doctrine that will underlie all future programs. COTS electronics is the underlying enabler: 1. Force multipliers: How to do more with less, while bring...


(Unofficial) electronic battlefield doctrine

I see three key tenets of DoD doctrine that will underlie all future programs. COTS electronics is the underlying enabler:

1. Force multipliers: How to do more with less, while bringing overwhelming lethality to the enemy. A unity of one asset, such as a Marine, can deploy individually but bring multiple assets to bear in real time.

2. Network-centric warfare: Connecting together as many battlefield assets as possible for information sharing and improved success;

3. Minimize U.S. casualties by relying on technology: This includes UAVs/UCAVs, standoff weapons, long distance C4ISR, and information warfare such as jamming and network data obfuscation (think deep packet examination and intentional data insertion or misrouting).

1996: COTS kickoff
The theme of this year’s Military Embedded Systems Resource Guide is “COTS: 10 years after.” Dr. William Perry’s famous “Perry Memo” was published in 1994, but it took about two more years of DoD and interservice arguing and hand wringing before the real weight of the memo sunk in: COTS wasn’t going away.

Back then, the SECDEF wasn’t changing his mind or policy. Commercial-Off-The-Shelf was going to be a mandate that the Services, primes, subs, and suppliers couldn’t dodge with fancy cost-plus contracts or funded overruns. Still, some contractors did try to fudge their way through contracts with sly wording that allowed them to design something COTS-like from scratch under federal funding. All too often these programs found themselves stuck with obsolete one-off technology, waivers to an Operational Requirements Document (ORD), and no budget to fix an in-house design.

In fact, as our panel of COTS executives points out (page 34, Spring 2006), the military is not going back to the pre-COTS days. Except for some highly specialized electronics applications such as space, high radiation, or unique military-specific sensors, the military just can’t revert to the pre-COTS days. “No way,” simply said one executive. 

While an increasing number of companies such as QP Semiconductor actively build MIL-STD-883 devices, the order of the day is to use even more consumer-quality COTS components, hardware, and software in modern defense systems.

2006: Latest COTS trends
That brings us to the present. The U.S. military and its allies rely heavily on electronics, specifically COTS electronics, and the capabilities of future military systems and weaponry can be spotted simply by reviewing today’s civilian and consumer electronics products and markets. What’ll be in your living room or coat pocket in two or three years? You’ll find that technology in mil programs 10 years from now.

“No way.” The military is not going back to the pre-COTS days.

At the recent Embedded Systems Conference in San Jose, California, I toured the aisles and met with vendors while keeping in mind what I call the DoD’s Electronic Battlefield Doctrine (see sidebar). After five days of dazzling displays, compelling demos, and some of the coolest embedded technology on the planet, here are some of my favorites and how they might apply to tomorrow’s military programs. I’ll be examining each of these in more quantitative detail in future MES issues.

Multicore processors: This is nothing new in military systems, but it’s taking AMD’s and Intel’s desktop CPUs to wake up the software market. Without the code, having extra hardware is useless. Intel’s Core Duo is the buzz in Apple’s new desktops and notebooks, and the software development ecosystem is really only starting. Most C4ISR and simulation apps migrate from Windows and Linux platforms, so multicore personal computers will have a huge impact on next-gen command centers and deployed x86-based systems.

Software optimization, verification, and embedded databases:The number of vendors offering these capabilities is just too large to list here (I stopped counting after about 20). Suffice it to say, as embedded systems get more complex and distributed, it’s becoming increasingly important to get the code done right. (And I’m not referring to DO-178B or MILS safety-critical code.) Also, look for distributed and embedded databases as battlefield assets seek to share data. Of particular note: the recent RTI announcement marrying NDDS publish-subscribe to Oracle’s Times Ten database.

Graphics-capable processors:Only the highest-res systems need dedicated graphics controllers anymore. General purpose CPUs can directly drive LCDs with reasonable performance and 2D video, even for moving maps and symbology overlay. The upside is smaller and lower power systems with ever more situational awareness powered by CPUs such as VIA’s new CX700.

PICMG’s Advanced Mezzanine Card (AMC) and PC/104’s EPIC: Look for these open standards-based hardware platforms to shoe-horn their way into more deployed defense systems. AMC cards are robust enough for many rugged apps, and PICMG’s MicroTCA cage/rack efforts may well offer an alternative to high-cost VME and CompactPCI systems.


For more information, contact Chris at [email protected].