in PDF Format - Embedded-Control-Europe.com

THINK OF THEM IN TERMS OF OUR POWER SUPPLIES 

Tiny, efficient, powerful: with their 

industry-leading digital control platform, 

Ericsson Power Modules’ new 

board-mounted power supplies really 

do prove that a little goes a long way in 

terms of power conversion efficiency. 

Today’s industrial and telecom systems 

use intermediate bus architectures requiring 

specialist skills to maximize the value 

of the relatively large amounts of power 

that they manage – which meant juggling 

power density against achieving the least 

possible conversion loss. A difficult task, 

for a conventional DC/DC converter using 

analogue control. 

The BMR453 DC/DC converter changes 

the game with its digital control platform. 

It allows an increase in power density 

from a typical 300W level to 400W in 

quarter-brick format, while offering a 

typical 96% conversion efficiency across 

its output power range. 

By harnessing the benefits of digital PWM 

and PMBus, more accurate control is 

possible in power critical applications, 

thus reducing consumption and CO 2 

emissions. The smaller and more efficient 

they are, the better they are for everyone. 

60 million. And counting 

Ericsson Power Modules has been a 

world-leading supplier of miniaturized and 

high-density DC/DC power modules for 

distributed power architectures since 1983. 

Since then, we’ve introduced just about 

every major first in the field, and lead the 

way in board-mounted product efficiency. 

Today, over 60 million board-mounted 

power supplies have been shipped to 

customers around the globe. We offer 

standard DC/DC converters and voltage 

regulators for information and communication 

technology equipment such as ADSL, 

MicroTCA, RF Power Amplifiers and many 

others; plus custom-designed solutions. 

Smaller size, greater efficiency 

The advanced design and efficiency of 

our products enables our customers to 

SOME PLACES YOU SHOULD START WITH: 

Ericsson AB; Power Modules Headquarters & Europe Sales Office: Phone: +46 10 716 96 20, 

e-mail contact: pm.info@ericsson.com. Germany, Austria, Switzerland: Josef Bose, Phone: 

+49 89 95 00 69 05. Italy, Spain (Mediterranean): Daniele Guidarelli, Phone: +39 33 55 78 63 91 

produce ever-more economically-viable 

designs, while reducing the energy 

consumption of their equipment. 

Wherever the quality and reliability of the 

power supply is crucial – from telecommunications 

equipment to the medical, 

avionics, military, space and industrial 

markets, our continued small improvements 

are bringing important benefits. 

Discover the benefits of using Ericsson 

digital power in your systems and learn 

more about the BMR453 – the digitally 

controlled DC/DC converter voted Product 

of the Year by Electronic Products. Meet 

us at Embedded World 2009 in Nuremberg, 

Germany – you’ll find the entire 

portfolio and our latest products waiting 

for you at the MEV booth, Hall 12-441. 

www.ericsson.com/powermodules

Dear Readers, 

At the moment the whole electronics 

industry is discussing which effects the 

financial crisis will have for the 

economy in general and especially for 

the electronics industry. Beside the fact 

that some semiconductor companies 

are struggling there are also good 

news. Kontron sales grew in 2008 of 

around 11% to approximately Euro 

495 million including a sales record of 

almost Euro 140 million in Q4 and 

even the order backlog is at a high level 

of over Euro 290 million. Siemens as 

another example has just published 

strong quarterly figures. The company raised sales in the first quarter 

of fiscal year 2008/2009 by 7 percent to 19.6 billion Euros, operative 

profit rose by 18 percent to 2.6 billion Euros.It seems that the situation 

isn´t generally that bad! 

Another positive indicator is the Embedded World 2009 exhibition and 

conference, which will take place from March 3 to 5 at the Nuremberg 

exhibition centre. The number of exhibitors has grown 14 percent to 

over 700 compared to last year. The conference will cover all aspects of 

embedded system design. In addition to hardware, software and tools, 

it will for the first time be devoting itself to topics such as green 

electronics or project management and thus picking up current trends 

and developments. 

Downturns force the embedded industry to develop new products and 

find new applications for these products. An example is described in our 

cover story. It shows the use of a single board computer in a patient 

monitoring system. From a functional point of view, such a system 

consists of a set of sensors, a central processing unit, human machine 

interface, a power supply with battery backup, and a logger and an 

alarm system. Using a single board computer as the core component, 

it is possible to quickly implement a basic monitoring system. The single 

board computer provides the interfaces that permit the implementation 

of a date logger, which makes it suited for the use in patient 

monitors. 

Another interesting article introduces INCA (industrial and network 

computing architecture), an open standards initiative for a sub-Micro- 

TCA platform designed to simplify the development of MicroTCAstyle 

systems, whilst reducing complexity and cost. Because there are 

some hurdles for using MicroTCA in industrial applications. MicroTCA 

is a complex system which consists of various components like power 

units, fans, backplane and chassis, system management, switch fabrics 

etc. INCA simplifies the system and merges for example the MCH and 

a PrAMC into a single blade called a PMCH, combining an application 

processor, system management and fabric switching. In INCA systems 

that do not have any hot-swap requirement, the power supply can 

power the backplane slots directly. In other cases where basic hot-swap 

or per-slot power control are required, then INCA supports a simplified 

management architecture. The result is a cost-effective system for 

industrial applications. 

Yours sincerely 

Wolfgang Patelay 

(Editor) 

VIEWPOINT 

3 February 2009 

V-1_2009-WOEI-4376 

Intel ® Atom 

meets 

COM Express 

Minimal power dissipation, external graphics and 

dual SDVO support are just some of the highlights 

you will find on MSC‘s embedded Intel ® Atom N270 

CPU design. 

Simply exchange the new MSC CXB-A945M COM 

Express modul against your current choice and 

participate on the new Intel ® ATOM technology: 

Up to 60% less power dissipation, low cost and 

identical interfaces – all with longevity support. 

� Intel ® Atom N270, 1.6GHz 

� Intel ® 82945GME 

� Up to 2 GByte SO-DIMM 

� Gigabit Ethernet 

� Dual SATA-300 

� Five PCI Express x1 

� PCIe x16 support 

� Dual Display support 

� HD audio 

� COM Express Type 2 

� Typ. 10W 

Tel. +49 8165 906 - 122 

boards@msc-ge.com 

Web Quick Link 1542-www 

Visit us! 

Nürnberg · 3. - 5. 3. 2009 

Hall 9 · Booth 235 

MSC Vertriebs GmbH 

Industriestraße 16 � 76297 Stutensee 

Tel. +49 7249 910-173 � Fax +49 7249 910-219 

� www.msc-ge.com

CONTENTS 

Viewpoint 3 

Medical Electronics 

Patient monitoring system with 

Atom-based single board computer 6 

Industrial Computing 

Deploying industrial MicroTCA 

with open standards and reduced cost 10 

Intel Atom processor enables 

low-power IPC for mobile applications 14 

Embedded World Preview 

Embedded World Product Highlights 16 

Exhibitors List – embedded world 2009 20 

Avionics 

Efficient and dependable real-time 

communications in UAV systems 24 

Implications of adopting aerospace 

development and verification standards 28 

Communications 

Platform management integrates 

hardware and software 31 

Small Form Factor Boards 

Pico-ITX single board computers 

combine advantages of COMs and SBCs 34 

45nm Core 2 Duo technology 

introduced in embedded COM modules 36 

Ruggedization 

Designing for wide temperature 

technology in industrial computing 38 

Customized CompactPCI CPU boards 

for industrial image processing 41 

Mainboard for long-term operation 

under extreme conditions in robots 43 

Product News 45 

Cover Photo 

EUROTECH 

February 2009 4 


Atom-based single board computer PAGE 6 

An interesting target application for the Proteus single board computer 

is patient monitoring systems. Such systems require various 

degrees of portability, with devices designed for use at the bedside, 

in fixed environments, or in completely mobile applications 

needing both powerful computational ability and flexible 

communications provision. 


with open standards and reduced cost PAGE 10 

This article introduces INCA (industrial and network computing 

architecture), an open standards initiative for a sub-MicroTCA 

platform designed to simplify the deployment of MicroTCA-style 

systems, whilst reducing complexity and cost. 


communications in UAV systems PAGE 24 

An increasingly wide range of 

unmanned air vehicles (UAVs) are 

posing unique design and development 

requirements particularly 

regarding the data link. Many 

defense application developers have 

turned to the open data distribution 

service (DDS) standard to provide the 

necessary messaging framework. 


hardware and software PAGE 31 

The new platform management 

service added by the Service 

Availability Forum provides the software 

with easy access to relevant 

state changes of hardware entities. 

This article explains its functioning 

with reference to a typical 

AdvancedTCA shelf with a number 

of blades represented in the information 

model. 

Pico-ITX single board computers combine 

advantages of COMs and SBCs PAGE 34 

This article introduces the new small form factor Pico-ITX for single 

board computers featuring the SUMIT interface which includes 

various future-proof busses. 


technology in industrial computing PAGE 38 

Wide temperature capability is a vital feature 

for industrial computers deployed in 

locations with harsh environmental conditions. 

This article explains the design technology 

needed for wide temperature 

industrial computers.

MEDICAL ELECTRONICS 


Atom-based single board computer 

By Pierfrancesco Zuccato, Eurotech 

An interesting target 

application for the Proteus 

single board computer is 

patient monitoring systems. 

Such systems require various 

degrees of portability, with 

devices designed for use at 

the bedside, in fixed 

environments, or in 

completely mobile 

applications needing both 

powerful computational 

ability and flexible 

communications provision. 

■ The Proteus SBC is based on the Intel Atom, 

a low-power X86 compatible CPU with a new 

internal architecture that provides significant 

improvements over previous generations, offering 

the same software environment and a 

performance approaching that available from 

desktops/notebooks, while requiring a power 

budget of only 12W. This result is achieved 

thanks to a totally revised architecture using 

45nm technology, and integrating numerous 

functions in the chipset, therefore eliminating 

the need for additional discrete chips. 

The introduction of the Intel Atom has been 

described as a revolutionary step, rather than an 

evolutionary one, since it provides a basis for 

the migration of applications requiring high 

performance processing into environments 

where low power consumptions and fanless operation 

are required. One crucial consequence 

is that this unifying process will also impact the 

area of software development, providing an 

infrastructure that allows common tools and 

applications to be used from the server to the 

embedded device. 

The Proteus SBC is offered as a complete platform 

(hardware coupled to pre-ported, readyto-run 

embedded operating systems) enabling 

very short product development cycles and 

corresponding time-to-market. It provides the 

full set of features embedded in the Intel Atom 

platform as well as extensive provisions for 

wired and wireless communications; any additional 

feature not included with the package can 

be easily implemented through two PCIe mini 

card sockets, or, in the case of the COM Express 

variant of the board, via a purpose-designed 

baseboard. Clearly, one of the most relevant aspects 

of the board is the optimization that keeps 

the overall power consumption (not including 

external devices and additional options such as 

PCIe expansion boards) within the 12W figure. 

This enables the development of battery-powered 

designs that deliver performance and 

extended usage time, opening a wide range of 

opportunities in mobile applications. 

It is worth noting that a number of advanced 

features are integrated by default; for instance, 

the chipset provides 3D graphics, with a hardware 

accelerator that releases the CPU from the 

intensive task of maintaining large-resolution 

displays. The single board computer actually 

provides two fully independent graphics channels 

with hardware HD video decoding which, 

coupled with sophisticated audio provision including 

integrated amplifiers, can simplify the 

design and reduce costs in a large number of 

target applications. 


It is interesting to consider some of these target 

applications in order to understand how the 

Proteus SBC can be effective in quickly implementing 

advanced, feature-rich products. One 

interesting application area is in patient monitoring 

systems. These devices are used to 

monitor the health of the patient, with internal 

applications that often involve the manipulation 

of multiple data sets using complex algorithms, 

together with the display of data in very 

precise detail. Such instruments require various 

degrees of portability, with devices designed for 

use at the bedside, in fixed environments, or in 

completely mobile applications needing both 

powerful computational ability and flexible 

communications provision. They are used 

whenever the subjects are in unstable conditions 

as the result of a disease or of a therapeutic 

treatment; there are also cases where it is 

important to log the physical parameters of 

healthy individuals, such as athletes. 

From a functional point of view, a patient monitor 

system consists of a set of sensors, a central 

processing unit, human machine interface, a 

power supply with battery backup, and a logger 

and an alarm system. Using the Proteus SBC as 

the core component, it is possible to quickly implement 

a basic monitoring system. Let us start 

with the sensors: a bedside system can be

Figure 1. Single board computer version of the 

Proteus 

equipped with modular units that provide the 

values of blood pressure, cardiac output, ECG, 

temperature and so on. These transducers and 

sensors can be managed using the onboard USB 

2.0 interfaces; designing USB sensor modules 

allows great expandability for the system and 

permits a gradual implementation of additional 

features. The computational capacity delivered 

by the Atom processor allows the analysis of the 

incoming data (filtering, pattern recognition, 

etc) and the interpretation of significant events 

that could require medical attention. The implementation 

of the human machine interface 

is greatly simplified since the SBC provides dual 

independent displays and touch panel support. 

SERIAL CONNECTIVITY 

INDUSTRIAL ETHERNET 

EMBEDDED COMPUTING 

WIRELESS & CELLULAR 

x86-Based Rackmount Embedded Computer with Modular Flexibility. 

� �� 

� �� 

� �� 

� �� 

� �� 

� �� 

� �� 

� �� 

Moxa. Industrial Networking Solutions. 

Moxa Europe GmbH | Einsteinstraße7|85716Unterschleissheim|Germany 

Tel.:+49893700399-28|E-Mail:Europe@moxa.com|www.moxa.com 

For instance, it is possible to implement a primary 

interface on a 15” UXGA panel where the 

operator can perform selections directly by 

touching the screen. An additional and independent 

(secondary) screen can be added on 

the second video channel, for instance on a remote 

link, to provide the operator with specialized 

information, remote control capabilities 

or visual feedback to the patient. Details 

such as screen orientation, resolution, colour 

depth are handled directly in the hardware, 

thanks to the integrated accelerator. For even 

more sophisticated data presentation and 

analysis, it is possible to use the hardware 3D 

accelerator which makes it possible to render 

complex scenes using high-level programming 

libraries, such as DirectX or OpenGL. Finally, 

the human machine interface also requires at 

least an emergency/help button and might include 

an optional mouse and keyboard. These 

features can be implemented with almost no effort 

using the GPIO, UART and USB interfaces. 

One important characteristic of the patient 

monitoring system is the power supply circuit 

that includes a battery backup system, thus providing 

uninterrupted service during power 

outages. The battery also doubles as the primary 

power supply when used as a mobile 

DA-680 Series 

x86-based Industrial 

Embedded Computers 

by 


device. These requirements are quite typical of 

a number of applications, so the Proteus has 

been designed to operate at +12VDC nominal 

and to accept an input range that spans from 

+8.5VDC to +25VDC. It is worth noting that 

the low power budget not only makes it possible 

to use a battery as the main power source, 

and hence permits the mobile use of the device, 

but also limits the heat generation. In the 

context of this example, a fanless design is very 

important since it greatly improves the reliability 

of the product and, at the same time, makes 

it possible to seal the unit, a very desirable property 

for a device that is used in an environment 

where fluids and other contaminants are quite 

common. 

A patient monitor might also require some sort 

of data logging, i.e. the patient parameters 

should be stored and made available at any time 

for medical or legal purposes. The SBC provides 

interfaces (one SATA and one MicroSD socket) 

that permit a straight implementation of a logger 

by adding a conventional hard disk or flash 

disk. Of course, due to privacy issues, some data 

might require encryption to prevent tampering 

and/or unauthorized access; implementing this 

feature is simplified by the availability of the 

on-board Atmel trusted platform module 

�� 

EMBEDDED WORLD IN NUREMBERG: 

March3to5,2009|Hall12|Stand450


Figure 2. Block diagram of the single board computer version of Proteus 

Figure 3. Application example: patient monitoring system 

(TCG v1.2), and the sheer computational performance 

of the Atom processor which permits 

the adoption of strong cryptographic algorithms. 

Additionally, if a printed output is 

required, the USB interfaces can be used to add 

a printer/plotter function. 

The final element in this example is the alarm 

system that must at least include some sort of 

acoustic feedback in case some event requires 

attention from the operator. The most basic implementation 

is a beeper that is triggered by the 

event; however, it is desirable to convey much 

more information, such as the relevance of the 

condition (warning, critical, etc) and its type 

(patient related, system fault, low battery, etc). 

While these messages can be delivered by a 

clever combination of tones, their interpreta- 

tion requires some training and might be a 

source of error in an emergency condition; 

therefore it is desirable to include speech synthesis 

or at least the capability of playing back 

prerecorded samples. The SBC provides advanced 

audio capabilities using Intel HD Audio 

with up to four audio streams (speaker L-R and 

independent headphone L-R) with an integrated 

2W audio amplifier on the speaker 

outputs, allowing these to be driven directly. 

In this scenario, the Proteus has made it possible 

to build a sophisticated mobile patient 

monitor with little hardware design efforts, 

replicating the features that are typically required 

in this type of device. Now, let us use the 

other features of the SBC to add advanced functions 

to the mobile patient monitor. The first 


enhancement is achieved by using the on-board 

Gigabit Ethernet interfaces: the first interface 

could be used to link the patient monitor to the 

hospital central server infrastructure where the 

patient records are stored. This would allow the 

operator to retrieve the clinical history of the 

patient at any time, and by using the touchscreen 

and optional keyboard/mouse, it would 

permit the submission of prescriptions and so 

on. The patient monitor could also stream the 

sensor data to the central server or to any other 

location (such as a monitoring room, where all 

patients can be simultaneously monitored 24/7 

byasmallnumberofoperators).Alternatively, 

the Ethernet interface might be used as a service 

port to provide remote technical support, 

such as to deliver software updates or to do 

remote diagnostics/inventory. 

Another interesting option could be implemented 

using one of the PCIe mini card sockets, 

where a WiFi/3G/GPRS communication 

adapter could be fitted. The wireless module 

could then provide services similar to the ones 

previously described for the Gigabit Ethernet 

and would make it possible to achieve the same 

degree of functionality even when the unit is 

used in the field (at home, in an ambulance or 

while the patient is transported from one room 

to the other). Another use could be to provide a 

backup link for the wired connections. 

The SBC also offers a GPS module option. The 

20-channel SirFStar III receiver, when coupled 

with the 3G (or GPRS, possibly with AGPS) 

adapter opens up a range of mobile services 

such as the geolocation of the patient, where a 

subject at risk, but otherwise not requiring hospitalization 

(e.g. for certain cardiac conditions) 

could be more effectively rescued in case 

of an emergency by sending a team in the right 

place at the early signs of a potential event, 

ideally even before the subject starts to realize 

that an emergency condition has developed. 

In an even more advanced scenario, the 

optional ZigBee (or Bluetooth) modules of the 

Proteus could be used to provide extra functions 

to the patient monitor. For instance, the 

ZigBee local network could create a contextaware 

environment, where the patient is recognized 

by the device, eliminating the need to 

enter any parameter or configuration. Similarly, 

the device could recognize the operator, 

activating only the functions that she is qualified 

to manage and logging the resulting 

who/what/when events. Another interesting 

feature that might be implemented with the 

ZigBee adapter is a totally wireless setup, where 

the patient is not tethered to the device, but 

wears a garment/smart shirt with embedded 

ZigBee-enabled sensors, therefore freeing the 

subject from the inconvenience of restricted 

movements. ■

There are a few things that you might not know...but need to know...about Aitech. First...we were first! 

That's right, Aitech built the world's first conduction-cooled mil-spec VME boards...years before there 

was NDI or COTS! And we are still way ahead of the pack in the COTS developments that matter most. 

We have COTS...complete! From advanced boards, enclosures/chassis, and OS and BIT firmware to fully integrated 

sub-systems and support services, including lifecycle and program management. For all COTS applications...land, sea, 

air...even space with 6U VME and 3U/6U CompactPCI and custom form factors! 

We have COTS...unique. Aitech has COTS products and capabilities that no one else has, including the highest 

MIPS/watt processor boards in the business; space-qualified, rad-tolerant COTS boards, multi-gigabyte mass Flash 

memory on a single board, high speed Ethernet boards, PMCs for PCM telemetry, A/D, D/A, Servos, and so much 

more...including AS9100 certification! 

We're global...and local. With world headquarters in California, multi-continental manufacturing facilities, and skilled 

representatives, tech support and agents throughout the world, Aitech can give you unsurpassed support and attention. 

We're ready when you are! There's a lot more about Aitech that you'd find very interesting. Give us a call or visit our 

web site. If you want COTS, get it from the company that invented it and still leads the way! 

From boards to sub-systems… 

we leave the systems integration to those 

who do it best – our customers. 

Aitech Defense Systems, Ltd. 

4 Maskit Street 

Herzlia, Israel 46120 

email: sales@rugged.com 

Tel: 972 (9) 960-0600 

Fax: 972 (9) 954-4315 

www.rugged.com

INDUSTRIAL COMPUTING 


with open standards and reduced cost 

By Peter Marek, Advantech 

This article introduces 

INCA (industrial and network 

computing architecture), 

an open standards initiative 

for a sub-MicroTCA platform 

designed to simplify the 

deployment of MicroTCA-style 

systems, whilst reducing 

complexity and cost. 

■ There are many good reasons to deploy MicroTCA 

which have been highlighted and discussed 

in great detail. The purpose of this article 

is to take a look at MicroTCA from a new 

perspective, indicate some hurdles and barriers 

in the way of its general industrial use, and introduce 

INCA (industrial and network computing 

architecture), an open platform initiative 

for a sub-MicroTCA platform. This is not a 

completely new approach, as in fact several vendors 

have already introduced custom sub-MicroTCA 

implementations that partially address 

the obstacles described below. So what is 

different about this article, and why is INCA 

different? 

There are two main answers. Firstly, INCA has 

been architected to become an open standard 

from the ground up. Only open standards with 

multi-vendor support can achieve critical mass 

in the short term, with economies of scale and 

technical evolution in the long term. Secondly, 

it goes beyond just specifying a low-cost AMC 

platform by taking a system-level approach and 

by introducing so-named lean mezzanine cards 

(LMCs). LMCs are simplified, AMC-like modules 

that serve two main purposes: supporting 

lower-cost implementations and reducing complexity. 

Overall, INCA has been designed to 

simplify the deployment of MicroTCA-style 

systems, whilst reducing complexity and cost to 

bring down engineering expense, COGS (cost 

of goods sold) and time-to-market. In the current 

economic environment the timing for this 

new system is absolutely right. Before taking a 

deep dive into INCA, let us have a look at MicroTCA 

as it stands today and summarize all its 

features and benefits, including however the 

hurdles that may be overkill and the barriers 

whichmaymakeitoverlycomplextodeployin 

industrial applications. 

The benefits of MicroTCA are as follows. MicroTCA 

is an open standard reusing AMCs. The 

ecosystem of 30+ vendors supplying AMCs and 

MicroTCA components and systems is alive 

and kicking. MicroTCA and AMCs are using 

serial interconnects such as GbE, PCIe, 

SAS/SATA and 10GE among others. Besides 

these popular serial interconnects, MicroTCA 

supports any kind of application-specific or 

custom protocol. The AMC-based form factor 

provides almost unlimited scalability and flexibility. 

Density is the highest amongst all modular 

computing standards. MicroTCA has been 

architected for five nines availability and full 

manageability, supporting component-level 

redundancy and hot swap. 

The hurdles on the way of using MicroTCA in 

industrial applications are these. MicroTCA 

makes complex systems. A MicroTCA system 

consists of one or more power units, one or 

more fan units, a backplane and chassis, and 


Figure 1. INCA reference system 

implementation 

one or two MCHs which take care of system 

management and provide the switch fabrics for 

the serial backplane interconnects. Power and 

fan units each have their own intelligence 

(management controller). All the elements 

mentioned are for basic system infrastructure 

support and perform no application processing. 

Although redundant power units or MCHs in 

a MicroTCA system are optional, redundancy 

has been designed into these components and 

to some extent into the system. The fully managed 

architecture with distributed management 

intended for high availability therefore comes 

with a cost overhead in applications with simpler 

management needs, or which do not require 

hot swap or redundancy. As always, flexibility 

is a gift for those people who know how 

to use it. 

For beginners, flexibility may be confusing. The 

best real world example was CompactPCI in its 

early days. Because of all the options and flexibility 

in CompactPCI, blade vendors were 

not sure which products to design for their customers, 

and customers did not know what features 

would be offered by standard products (a 

classical chicken-and-egg scenario). A little of 

the same goes for MicroTCA as well, just a little 

though: the many options to choose data 

plane (fat pipes) switch fabric interconnects 

complicate architecting the system, and make 

component-level selection difficult, especially

Figure 2. Advantech UTCA-5500 PMCH integrates MCH and a PrAMC 

for backplanes and MCHs. The barriers against 

using MicroTCA in industrial applications are 

as follows. With high availability built in, MicroTCA 

demands some higher-level management 

usually referred to as the system manager. 

In the telecom world, such managementplane 

software typically runs on top of high 

availability middleware based on HPI on an 

x86-based processor module, or even from a 

remote part of the network. In applications 

where requirements for high availability are 

lower or where simpler redundancy models are 

sufficient (enterprise applications often use 

box-level rather than component-level redundancy), 

or where neither redundancy nor hotswap 

are needed, then high availability represents 

a pure overhead to the system integrator 

and user. Even though MicroTCA can be de- 

Figure 3. Implementation of a simple IO function on an LMC (top) vs. AMC 


ployed without such high-level system management, 

the concepts behind it have been built 

into the systems and still need to be understood. 

The main reason is because management in MicroTCA 

is distributed to different entities 

(power unit, fan unit, MCH and x86 processor 

AMC) each of which is usually sourced from 

different vendors (that being the idea of an open 

standard and ecosystem). Although the MicroTCA/AMC 

ecosystem is making great efforts 

to ensure interoperability, both in terms of specification 

work under PICMG and real-world 

testing, the interfaces between each entity must 

still be understood by the system integrator. 

Custom hardware migration/integration is perhaps 

the biggest obstacle for wider industry MicroTCA 

adoption. Custom hardware refers to 

third party hardware which is not yet available 

in an AMC form factor, or to proprietary and 

application-specific hardware. Usually, combined 

with specific software, such hardware is 

key to implementing the overall system functionality. 

Such intellectual property is typically 

never outsourced even when OEMs move 

away from proprietary platforms to openstandards-based 

hardware and software. The efforts 

required to migrate these company crown 

jewels are one of the main criteria for platform 

selection. 

MicroTCA and AMCs do not do a good job 

here. No matter whether hot swap or advanced 

system management is needed or not, OEMs are 

forced to understand how to design an AMC 

which supports these features in order to migrate 

their hardware to MicroTCA. This includes 

IPMI-based management in the form of 

management controller hardware, associated 

firmware and support for both hot swap and Ekeying. 

IPMI needs to be designed regardless of 

whether the module needs to be managed or 

hot-swapped in the final application. Unfortunately, 

MicroTCA system management is based 

on complex specifications which do not help to 

simplify this task. The MicroTCA specifications 

reference the AMC specifications, AMC in turn 

references ATCA, and ATCA references IPMI 

V1.5. So designing and debugging a module that 

complies with all of that represents a higher level 

of complexity compared to integrating off-theshelf 

AMCs that have been designed to those 

standards and have gone through reasonable interoperability 

testing. 

The objectives of INCA can be simply stated as 

getting rid of the hurdles and the barriers and 

keeping as many as possible of the benefits. Let 

us take a look at the basics. INCA merges the 

MCH and a PrAMC into a single blade called a 

PMCH, combining an application processor, 

system management and fabric switching. A 

PMCH offers GbE and PCIe switching and 

SATA ports. It introduces simple plug-and-play 

11 February 2009


Figure 4: Block diagram of a PMCH 

USB connectivity and wide PCI Express links 

up to PCIe x16, to support industrial graphics 

or very high speed IO. The PMCH takes care of 

high-level and low-level system management 

thereby simplifying the system infrastructure 

and reducing costs. Actually the centralization 

of system management on the PMCH makes it 

possible to reduce the complexity visible to the 

integratoranduserorevenhideitawaycompletely. 

An INCA system can host up to 12 

AMCs or LMCs (the same as MicroTCA). 

LMCs are IO boards which are passively managed 

by the PMCH and basically support the 

fabrics offered by the PMCH. With PCIe, GbE 

or USB-based connectivity, the most popular 

Feature 

Management 

Hot swap 

Redundancy 

Fabrics 

E-keying 

PCIe 

USB 

LMC 

LMCs support a simple ID 

(FRU) EEPROM and a 

temperature sensor connected 

to PMCH over I2C 

Only support for basic hot swap 

requiring manual interaction by 

user 

No support 

LMCs support GbE, 

SAS/SATA, PCIe and USB. 

Other fabrics can only be 

supported as local links to other 

module slots on the backplane. 

E-keying is performed by the 

PMCH on a module basis. Only 

if all the LMC fabrics comply to 

the connected counterparts in 

the system, will the module be 

powered. 

Up to PCIe x16 

Supported 

and well understood interconnects are supported 

making it fairly easy to design LMCs or 

migrate legacy designs based on parallel busses 

such as PCI and ISA. In particular, as there is no 

requirement to understand and implement 

IPMI, designers can focus their effort on the 

core functionality. 

INCA does not require specially designed, intelligent 

power and fan units. Off-the-shelf, unmanaged 

fans and power supplies simplify the 

overall system design and help to reduce cost 

(especially when it comes to AC-powered 

systems). If needed, fans can be managed by the 

PMCH over I2C. In INCA systems that do not 

AMC 

AMCs implement an IPMI-based 

module management controller (MMC) 

that connects to the MCH/PMCH 

over IPMB-L 

Full hot-swap support 

Supported by μTCA 

AMCs can support any fabric. 

E-keying is performed on a per-port 

basis by the MCH/PMCH. Only 

matching ports will be enabled. 

Although not specifically limited by the 

AMC specification, AMCs usually just 

support PCIe x4 links. 

Not supported 


have any hot-swap requirement, the power supply 

can power the backplane slots directly. In 

other cases where basic hot-swap or per-slot 

power control are required, then INCA supports 

a simplified management architecture 

where the PMCH controls hot-swap switches 

on the backplane. It is quite obvious that this 

helps to reduce system complexity and cost as 

well. Power-management-related costs can be 

minimized for those systems which do not need 

it and for those which do, costs will scale with 

system slot count. 

Introducing USB further reduces overall system 

cost as well as simplifying migration to the 

INCA platform. USB is well understood and offers 

an excellent price/performance ratio. USB 

silicon is cheap and widely available. There are 

a lot of ready-to-use peripherals that support a 

USB host interface that can be mounted on a 

LMC. As an example, in figure 3, imagine what 

effort it takes to implement some general purpose 

inputs and outputs on MicroTCA as 

compared with INCA. 

On MicroTCA you will first of all need to implement 

an AMC and consequently need a 

management controller for IPMI-related 

firmware. Next, you will probably choose PCIe 

as host interface and will need a bridge chip or 

FPGA to implement the host interface. Finally 

you will implement a small microcontroller to 

drive and sample your general purpose IO pins. 

Conversely, on INCA, you build an LMC that 

has an ID EEPROM for basic management support 

and a small microcontroller to drive and 

sample the general purpose IO pins using 

USB as a host interface. That small microcontroller 

probably does not cost you more than 

the management controller required on an 

AMC and used just for IPMI. Furthermore, 

think of how to add serial ports to a system. The 

least expensive way is to use USB-serial converter 

chips with drivers for all popular operation 

systems ready to go. 

In INCA, the PMCH and management architecture 

support devices which do not have to be 

installed in backplane slots, in particular on 

SATA and USB ports. This opens the door to 

easily integrate USB-based peripherals such as 

card readers, wireless modems and many others. 

This is especially interesting if those peripherals 

are larger than an AMC/LMC or come 

packaged and pre-certified according to international 

regulations. AMCs and LMCs can only 

support 2.5” disk drives because of physical limitations, 

so by supporting drives mounted 

elsewhere in the system the related capacity barriers 

can be by-passed and costs reduced even 

further. The PMCH can be implemented in a 

single or double-wide AMC form factor. A 

block diagram is shown in figure 4. The PMCH 

may have two backplane connectors, named P1


and P2, corresponding to connector 1 and connector 2 of an MCH. P1 

is mandatory and carries the interfaces for basic system management 

as well as fabrics for up to 6 AMCs/LMCs. P2 is optional and carries the 

fabrics for additional 6 AMCs/LMCs: P1 contains power, IPMB-L, hot 

swap, management IOs, I2C incl. ALARM, 6 GbE ports, 2 SATA ports, 

4 USB ports, 8 PCIe lanes, including 5 clocks, 2 PCIe x4 or 1 PCIe x4 

+ 4 PCIe x1. P2 contains: 6 GbE Serdes links, 2 or 4 SATA ports, 4 or 

0 USB ports, 16 PCIe lanes, including 4 clocks, 4 PCIe x4 or 2 PCIex 8 

or 1 PCIe x16. 

The MMC on the PMCH will take over low-level management functions 

whilst high-level management can be executed on the x86 processor. 

Sharing system management and application processing on one 

processor may look problematic at the first glance. Consider, however, 

that in INCA-related application scenarios the bulk of system management 

only needs to be performed at system power-on before applications 

are up and running. When the system is up and running, processor 

loading due to system management is usually negligible and principally 

related to thermal management. Events creating higher loading 

for system management usually represent some critical problem where 

a degradation of application performance is acceptable. To eliminate dependencies 

between system management and application software, virtualization 

may even be used on the latest multi-core processors. 

The nice thing about LMCs is that they are fairly straight forward to implement 

and AMCs can be operated in the same backplane slot. Table 1 

highlights implementation differences between AMCs and LMCs. ■ 

Product News 

■ Kontron: basic motherboards with 45nm Intel 

quad-core processor 

Kontron has extended its family of basic motherboards with two 

high-performance variants based on the 45nm Intel Core 2 Quad 

processor: The Kontron KTG41/ATX ATX basic motherboard and the 

Kontron KTG41/ATXU Micro-ATX basic motherboard. Compared 

to standard ATX and Micro-ATX motherboards the two new Kontron 

ATX and Micro-ATX basic motherboards with Intel G41 Express 

chipset and LGA 775 socket for Intel processors up to the 45nm Intel 

Core2 Quad processor Q9650 offer advanced design qualities for 

rugged environments plus 3 years long-term availability and vendor 

support for OEMs. Equipped with only the latest and most demanded 

interfaces, Kontron basic motherboards are extremely cost-effective, 

making them ideal for highvolume applications with fast innovation 

cycles in the fields of gaming, digital signage, POS, check-in terminals, 

ticketing machines, hotel multimedia terminals or even industrial 

shop floor applications for quality control. 

News ID 750 

■ N.A.T.: power saving AMC module for industrial automation 

N.A.T. introduced its newest member of the family of AMC modules. 

Based on a PowerQUICC III the NAMC-8560-IO meets the performance 

and power consumption requirements of customers in the 

industrial automation market, wanting to benefit from the advantages 

of MicroTCA. The NAMC-8560-IO is equipped with an MPC8560 

and provides 256 MByte memory, 32 MByte Flash. Additionally to a 

PCIexpress interface and a Gigabit-Ethernet interface at the backplane, 

another Gigabit-Ethernet interface is available at the front panel. Thus 

connections to cameras or EtherCAT-IO can be easily realized. 

News ID 727 

13 February 2009


Intel Atom processor enables 

low-power IPC for mobile applications 

By Christian Lang, DSM Computer 

The low power consumption 

of the Intel Atom processor 

enables fanless embedded 

PCs in closed housings, suited 

to mobile applications, with 

battery life extended two to 

three times, opening up completely 

new possibilities. In 

combination with the Intel 

82945GME chipset, a wide 

variety of important industrial 

interfaces can be offered. 

■ In recent years, the focus has increasingly 

shifted to the energy efficiency of embedded 

computing systems, in addition to computing 

power. The growing number of industrial applications 

in which fanless industrial computers 

(IPCs) are realized with corresponding I/O 

performance is driving this trend. Low power 

consumption reduces the waste heat of the systems 

and lowers demands on the cooling concepts. 

In this way, completely fanless designs can 

be constructed in closed housings. Thanks to 

low power consumption, the power-saving 

embedded PCs are also suited to mobile applications 

and enable an extension of battery life 

by two to three times. This opens up a new 

range of applications for these IPCs that only a 

matter of months ago seemed to be impossible. 

A prerequisite for the development of embedded 

systems with extreme power savings is the 

Atom processor architecture, newly developed 

by Intel, with unparalleled performance characteristics. 

The processor is manufactured in a 

45nm process and is cased in small form factor 

housing (SFF). The chip, which is just 

13x14mm² or 22x22mm², depending on the 

version, takes up very little space on the PCB 

and features especially low power dissipation. 

The majority of the embedded computing 

platforms presented in recent weeks, based on 

the power-saving Intel Atom processor archi- 

tecture, integrate a CPU from the Z5xx series 

and the Intel US15W system controller hub. 

The two-chip solution is persuasive as a result 

of its high energy efficiency, but does not 

however offer the wide range of modern interfaces 

required by many applications, in particular 

in industrial automation technology. 

Therefore, only two PCI Express Lanes are 

available, of which one will normally already be 

used for the Gigabit LAN interface. Additional 

interfaces such as further PCI Express or a PCI 

bus can only be realized via special bridge chips. 

A solution is offered by the combination of the 

power-saving Intel Atom N270 processor, the 

82945GME Northbridge and the ICH7 Southbridge. 

The Intel 82945GME chipset offers a 

wide variety of interfaces important for the industry, 

such as a PCI Express x16 interface for 

external high-end graphics cards and a dual 

DVI/LVDS interface. Additional system expansions 

are possible via five PCI Express x1 

Lanes. The combination of the Intel Atom CPU 

and Intel 945GME chipset makes much more 

sense than many of the CPU boards on the 

market with the Intel 945GC chipset, which exhibits 

8 to 9 times the power dissipation of the 

processor. The Intel Atom N270 processor is 

clocked with a frequency of 1.6GHz, and with 

a thermal design power (TDP) of 2.5W has extremely 

low power consumption. The complete 

three-chip solution has a theoretical power dis- 


Figure 1. The first 

embeddedsysteminthe 

NanoServer family based 

on the energy-saving 

Atom N270 processor 

from Intel 

sipation of 12.6W. In this way, a range of fanless 

applications can be realized that previously only 

used the generally expensive low-voltage or 

ultra-low-voltage processors. The N270 CPU 

supports Intel Enhanced SpeedStep technology. 

As a result of the integrated hyper-threading 

technology, multiple threads can be executed at 

the same time in a single core. 

DSM Computer utilizes the advantages of the 

high energy efficiency of the Intel Atom processor 

technology, and the associated lower system 

costs, in its new E3-COM NanoServer embedded 

PC. The industry-compatible NanoServers 

can be used as distributed and independent 

client servers, for example as control unit, 

measured value or communications servers. 

The core of the E3-COM NanoServer is a COM 

Express baseboard, developed for this application, 

in MiniITX format with a COM Express 

module based on the N270 CPU and the Intel 

82945GME chipset. In this way, DSM is placing 

emphasis on the long-term use of the globally 

recognized COM Express PICMG standard, 

which ensures the long availability of the individual 

components on the Intel embedded 

roadmap. The baseboard and the CPU module 

are manufactured in Germany. The scalable 

COM Express module utilizes the benefits of 

the Intel Atom processor due to its small size of 

125mmx95mm (basic format). In addition, the 

interfaces can be led to the outside via the in-

Figure 2. Customer-specific basis board with 

energy-saving COM Express module 

Figure 3. The E3-COM NanoServer has a 

wide range of interfaces. 

dividual carrier board in the place where they 

are required in the housing design. Therefore, 

the external connections do not have to be cabled 

in a complex way in the housing. COM Express 

defines all specifications relevant for the 

industrial application such as PCI Express, PCI, 

PEG, USB 2.0, Gigabit Ethernet and SATA via 

the two pin-out specification types 1 and 2. The 

COM Express connector offers high shock 

and vibration stability as well as good EMC behavior. 

The E3-COM NanoServer is accommodated in 

a high-quality industrial housing with compact 

dimensions of 220x220x110mm³. There are 

four USB 2.0 ports for the flexible connection 

of various peripheral devices and high-speed 

Gigabit Ethernet. There are four further USB 

2.0 ports on the baseboard and these can be led 

out of the housing if required. The required 

audio functionality can be ensured via the Intel 

high definition or the AC'97 audio interface. A 

fast PCI Express 1x Lane as well as a PCI bus 

(3.3V / 5V) and a MiniPCI bus complete the 

picture. Additional interfaces can be realized on 

a customer-specific basis via the I/O shield of 

the embedded system. 

Thanks to the 950 Graphics Media Accelerator 

integrated into the Intel 945 chipset, the IPC offers 

excellent 3D graphics features. The VGA 

memory has a capacity of 224 Mbytes (shared). 

The internal graphics supports dual independent 

displays. If required, the graphics capability 

can be expanded via the PCI Express Graphics 

(PEG) port. A backlight interface can also be 

led to the outside in addition to the VGA, DVI- 


D and LVDS interfaces. With the assistance of 

the backlight control, the light intensity for the 

backlighting of the LCD display module of a flat 

screen can be controlled. In order to ensure that 

the panel can be operated in energy-saving 

mode, the backlight is switched off after a prescribed 

interval of time. Displays that are 

equipped with a continuously working backlight 

control provide constant grayscale display. 

This function is on the baseboard and can also 

be led to the outside in a customer-specific variant. 

For data storage, there is space in the solid PC 

housing for a rugged 2.5 inch hard disk, which 

is suited to the temperature range of -20°C to 

+75°C and therefore works reliably at low external 

temperatures. The robust 80 Gbyte hard 

disk can run round the clock without interruption, 

reaching an MTBF of 750,000 hours. 

As an option, connection of a disk-onmodule 

or compact flash is possible. 

The IPC runs on the Windows XP operating 

system (embedded). The TrustedCore BIOS 

from Phoenix Technologies is implemented on 

the COM Express module, which includes security 

functions corresponding to the Trusted 

Computing Group (TCG). This means that the 

copying and modification of customer knowhow 

is prevented. For the optimization of the 

BIOS, DSM has full access to the Phoenix 

source code. In addition, the Computer-on- 

Module integrates the Infineon Trusted Platform 

Module TPM 1.2 SLB 9635. 

The target market of the E3-COM embedded 

system from the NanoServer family, based on 

the Intel Atom processor, is mainly mobile applications, 

for example in the areas of industrial 

automation technology, transport and logistics, 

medicinal devices and POS & POI systems. A 

new application of the compact computer is 

digital signage, which identifies a networked audiovisual 

marketing and information system, 

the content of which can be very rapidly programmed 

in a targeted way using software. Digital 

signage is used for example at rail stations 

and airports in order to show departure and arrival 

times on a large-scale display. News and 

advertising are also targeted to specific groups 

in large hotels, clinics and well-visited shopping 

malls. Combining these with, for example, 

event details and local weather information assures 

that more customer attention is secured 

than in the case of pure marketing displays. The 

task of the NanoServer in this application is primarily 

to control the marketing display with the 

representation of the contents. For applications 

requiring very high computing power, the E3- 

COM NanoServer with Intel Core 2 L7400 duoprocessor 

is also supplied. This system can be 

operated in the extended operating temperature 

range of -10°C to 50°C. ■ 


boards. systems. solutions. 

Embedded Computers 

Made in Germany. 

www.ekf.com 

CompactPCI ® 

(Express and more …) 

with CCM-BOOGIE 

• Intel ® Core 2 Duo up to 2.26 GHz 

• Up to 8 GB DDR3 RAM (4 GB soldered) 

• Dual Screen Graphics 

• 6 x SATA, 11 x USB 2.0 

• Versatile Mass Storage Options 

(CF, HDD, SSD) 

• Rear I/O and Mezzanine Expansion 

via PCI Express 

• Customized Expansion Boards 

• Ruggedized and long time available 

Besuchen Sie uns: 

Embedded World 

03.–05. Feb. 2009 

Halle 12, Stand 313 

EKF Elektronik GmbH 

+49 (0) 2381 68900 

www.ekf.com 

sales@ekf.de 

Your partner since 1972.

EMBEDDED WORLD PREVIEW 

Embedded World 2009 Preview 

Despite the severe economic 

downturn, it seems 

that Embedded World, 

the No. 1 event in the 

embedded market, 

is continuing its 

record-chasing run. 

The number of exhibitors at Embedded World 

has grown 14% against last year to more than 

700. But we will have to wait and see, whether 

there will be a new visitors record too. Embedded 

World will take place from March 3 to 5, 

again at the Nuremberg exhibition centre. 

Visitors can register online at 

www.embedded-world.de in advance for their 

visit to the fair thus securing quick and free admission 

to embedded world 2009 at the event. 

The embedded world Conference will once 

again cover all areas of embedded system 

Hall-Stand 12-318 

Schroff: brochure about everything an 

enclosure system needs 

From non-standard system solutions, modified 

off-the-shelf components to individual developments. 

In a 16-page brochure Schroff has 

presented its products, services and competencies 

for developing and manufacturing co-ordinated 

electronic systems. The brochure outlines 

the various standard components of a system, 

on the basis of which specific systems can 

be configured. With its professional quality 

management, certification and conformity to 

standards, Schroff meets the conditions necessary 

for ensuring the security and dependability 

of systems and their individual components. 

News ID 17 

development. In addition to hardware, software 

and tools, the conference will for the first time 

also be devoting itself to themes such as green 

electronics or project management and thus 

picking up current trends and developments. In 

their lectures, experts and specialists from the 

embedded community will provide answers to 

the most pressing matters and ensure that the 

embedded systems developers are best prepared 

for the challenges of tomorrow. 

The electronic displays Conference is already 

entering the third phase in Nuremberg. The 

conference has been able to establish itself 


Concurrent: 3U CPCI board with 

1.6 GHz Z530 processor 

Concurrent Technologies introduces their latest 

ultra low power 3U single/dual slot CompactPCI 

single board computer, the TP 

A40/30x. The board uses the 1.6 GHz Intel 

Atom processor Z530 and the highly integrated 

Intel System Controller Hub US15W both 

from the Intel embedded roadmap, ensuring 

long term availability. With 1 Gbyte DDR2- 

533 SDRAM this ultra low power board also 

supports a variety of peripheral I/O ports, 

CompactFlash, CANbus controller, an optional 

PMC/XMC module and can operate in a system 

slot, peripheral slot or as a blade. 

News ID 716 


parallel to embedded world surprisingly quickly 

– in 2008 over 50 papers were presented by 

more than 200 participants. 

Professor Dr. Karlheinz Blankenbach, 

Hochschule Pforzheim (Technical College), 

Chairman of the Congress Advisory Board, is 

confidently looking ahead to 2009: 

“Following successful electronic display 

Conferences in 2007 and 2008, the Advisory 

Board carried out some fine tuning to several 

areas of the event in March 2009: as a result, this 

time the sessions will be followed by “Author 

Interviews”. 


VIA: fanless touch-screen panel PC 

VIA announces the VIPRO VP7710 touch 

screen panel PC, a full featured, entirely fanless 

panel PC suitable for intelligent display applications 

in demanding environments. Rugged, 

stable and flexible, the VIA VIPRO is the first in 

a series of advanced display technologies from 

VIA. Amid growing public acceptance of intuitive 

touch screen technologies, the VIA VIPRO 

addresses an increasing demand for cost effective, 

intelligent displays in commercial applications 

such as ticketing, ATM, vending and information 

kiosks as well as sophisticated fleet 

deployment infrastructures in transport, delivery 

and logistics enterprises. 

News ID 788

Hall-Stand 11-306/12-404 

Kontron and LynuxWorks: safety 

critical platform 

Kontron and LynuxWorks released an Intel 

based COTS platform for safety critical, deterministic 

real time embedded applications, 

using the Kontron PENTXM2 single board 

computer running the LynxOS-178B RTOS. 

The Kontron PENTXM2 uses the 1.67 GHz 

dual-core Xeon, Intel's advanced low-power x86 

technology, combined with the Intel E7520 

server class memory controller hub. The PEN- 

TXM2 is available with up to 4GB of DDR2- 

400 SDRAM. When paired with the support of 

VITA 31.1 backplane networking, the PEN- 

TXM2's VITA 38 intelligent platform management 

interface feature provides for easy scaling 

into a multiprocessing system. 

News ID 898 

Hall-Stand 9-477/12-142/12-314/12-306 

Fujitsu Siemens: mainboard controls 

KUKA robots 

To operate the PC-based control system of their 

robots, KUKA works with Fujitsu Siemens 

Computers’ motherboard D1688-K. KUKA 

has selected the FSC industrial mainboard for 

its performance, consistent high quality and its 

price as well as the guaranteed long-term avail- 

ability of all components. Special emphasis was 

placed on reaction times (track planning, 

graphical display, network communications, serial 

interface communications) and access 

speed to memory components or PCI cards 

(sensors, field busses, inputs/outputs). 

News ID 857 

Hall-Stand 9-429/12-565 

Advantech: rugged compact 

Embedded box computer 

Advantech has just released its newest Compact 

Embedded Computer ARK-1382, the smallest 

media capable compact embedded computer in 

the ARK 1000 series. ARK-1382 delivers excellent 

computing and multi-media performance from 

its Intel Celeron 1.06 GHz processor and Intel 945 

GM chipset. Processing and multimedia performance 

are high on ARK-1382’s list with dual 

DVI-I independent displays, each capable of supporting 

a 16:9 aspect ratio wide screen display. 

News ID 793 


Green Hills: INTEGRITY achieves EAL6+ 

security certification 

Green Hills announces that the INTEGRITY- 

178B operating system has been certified by the 

NIAP to Common Criteria Evaluation Assur- 


ance Level (EAL) 6+, High Robustness. This 

certification is the highest Common Criteria security 

level for an operating system. Only an 

EAL6+ High Robustness operating system is 

certified to protect classified information and 

other high value resources at risk of attack from 

hostile and well-funded attackers. This is secure 

by anyone’s definition. 

News ID 813 


Eurotech: industrial GSM/GPRS/3G 

wireless router 

Targeted for telemetry, Internet Access and Mobile 

Computing applications, the rugged 

ZyWAN from Eurotech offers a secure and robust 

connection across GSM/GPRS/3G, CDMA 

mobile data networks. It provides real-time access 

to any Ethernet, 802.11b/g or serial device 

for mobile and fixed data applications. The 

ZyWAN operates as a fully configurable embedded 

Linux router enabling firewalling, 

DHCP, DNS and NAT. Also included is a GPS 

receiver providing both a sophisticated tracking 

program and NMEA data strings for mapping 

applications. To easily manage the ZyWAN, a 

WEB GUI interface presents a simple tool to 

quickly change settings locally or over-the-air. 

News ID 911


Hall-Stand 11-118/10-239 

Wind River: VxWorks MILS platform 2.0 

in evaluation 

Wind River announced that Boeing, Lockheed 

Martin, and Raytheon are among the 

companies currently evaluating early access releases 

of VxWorks MILS Platform 2.0. The 

MILS (multiple independent levels of security) 

infrastructure enables development of multilevel 

secure systems that process classified data 

at all levels simultaneously, including downgraders, 

guards, and other cross-domain applications. 

The VxWorks MILS Platform will 

enable multiple teams working in parallel to 

build the systems. 

News ID 806 

Hall-Stand 11-F114 

LynuxWorks: hypervisor for high 

assurance systems 

LynuxWorks announces the availability of the 

LynxSecure 2.0 next-generation separation 

kernel and embedded hypervisor for high assurance 

systems. LynxSecure provides a standards-based, 

secure, multi-domain computational 

platform that serves as a trusted foundation 

for applications with low, medium and 

high assurance requirements that are all running 

concurrently on a single system. It allows 

legacy applications to run unmodified, enabling 

systems to be modernized with increased 

information sharing and security. 

News ID 801 

Hall-Stand 11-306/12-404 

Kontron: hardware platform for 

Märklin Central Station 60213 

At the SPS/IPC/Drives Kontron has presented 

the hardware platform for the new Central Station 

60213 from Märklin Digital. The platform, 

which Märklin introduced in September 2008, 

offers more application flexibility than any 

other known control system for model railways. 

It communicates via Märklin’s own interfaces as 

well as numerous standard interfaces such as, 

for example, USB and Ethernet and repesents a 

new milestone in the control of complex model 

railways. 

News ID 856 


ELMA: intelligent active backplane 

is MicroTCA compatible 

ELMA has integrated the management functions 

of the power module onto the backplane 

with ELMA’s new blu!eco basic system. The 

active backplane thus plays a substantial role in 

reducing system costs and the advantages of 

forward-looking MicroTCA technology can 

now be used economically for a variety of automation 

tasks. blu!eco is fully compatible 

with MicroTCA specifications. blu!eco uses a 

tried-and-tested, pluggable ATX power supply 

with +12 V/300 W. The IPMI ‘intelligence’ for 

the voltage supply, hot-swap functionality and 

fan controller are, on the other hand, all integrated 

onto the backplane. 

News ID 355 


EKF: mezzanine CompactFlash 

adapter modules 

Due to their reliability, industrial grade CompactFlash 

memory cards are very popular for 

rugged embedded applications, as replacement 

or in addition to hard drives. Available for the 

EKF series of CompactPCI CPU boards, three 

mezzanine CompactFlash adapter modules 

have been designed for either front loading of 

the CF media, or fixed mounting directly on top 

or bottom of the carrier board. The C13 

adapter module allows easy insertion of a 

CompactFlash card, from a slot in the 6HP 

front panel. 

News ID 25 


Axiomtek: slim touch panel computer 

with AMD LX800 

Axiomtek has introduced the GOT-5840TL, an 

8.4’ super-slim touch panel computer with only 

44mm in thickness, which supports a low 

power AMD LX800 500 MHz processor with 

fanless operation. The GOT-5840TL adopts a 

rugged NEMA 4 / IP65 compliant plastic front 

bezel, and is equipped with an 8.4’ SVGA TFT 

LCD with resolution up to 800 x 600 pixels. 

This super-slim touch panel computer is 

equipped with one MiniPCI slot for expansion 

availability, and has a CompactFlash card for 

storage capability. 

News ID 18 


DSM: small fanless control computer 

for POS/POI 

With its small dimensions of only 180 x 180 x 

55 mm, the PicoBox by DSM Computer is a 

fanless control computer for POS/POI and digital 

signature applications. The PicoBox augments 

DSM's embedded systems product line 

with the high-performance NanoServers and 

More information about each news is available on 

www.Embedded-Control-Europe.com/bs_magazine 

• You have to type in the “News ID”. — 


the scalable BookSize PCs for ultra-compact applications. 

The PicoBox is currently available in 

two variants. The first variant is the PicoBox C2 

(96M1611L) model based on the low-power 

Intel Atom N270 processor with a clock frequency 

of 1.6 GHz. The integrated Intel 

82945GSE chipset with the Southbridge 

ICH7M also operates very energy-efficiently. 

News ID 886 


MEN: VMEbus PC used in CERN particle 

accelerator 

The new VMEbus SBC from MEN Mikro 

Elektronik will be used in CERN’s particle accelerator. 

The A20 supports a multitude of Intel 

Core 2 types and pairs outstanding reliability in 

computing with fast data transmission. The 

A20 single-board computer is the latest 6U 

VMEbus board of MEN's product range. 

News ID 860 


Curtiss-Wright: 3U VPX card for networking 

Curtiss-Wright has introduced its highest density 

small form factor Gigabit Ethernet multilayer 

switch/router board designed for rugged 

embedded aerospace and defense applications. 

The VPX3-683 FireBlade, available with 24 GbE 

SerDes ports and up to two x10 GbE XAUI 

ports, is for system integrators architecting secure 

high performance IPv4/v6 intra-platform 

networks. This rugged, compact 3U VPX card, 

which can operate either as a fully managed 

switch/router, provides significant performance 

and configuration advantages to developers 

building layer 2 and layer 2/3+ networks. 

News ID 55 


AAEON: economic industrial motherboard 

with VIA C7/ Eden processor 

AAEON unveils the IMBM-700, an industrial 

motherboard using the VIA C7/ Eden processor. 

Specially designed to support existing 

equipment in the Factory Automation market, 

the IMBM-700 offers legacy interface support 

for the ISA bus. The IMBM-700 is based on 

VIA’s CN700 and VT8237R+ chipset and supports 

both the C7-M Ultra Low Voltage 1.0GHz 

for extremely low power applications all the way 

up to the C7-M 2.0GHz for applications requiring 

higher performance while maintaining 

low power requirements. Two 240-pin DDR II 

400/533 DIMM sockets support up to 2GB. 

News ID 947


Concurrent: single-slot cPCI SBC with 

dual-core processor 

Concurrent Technologies has announced the 

release of their latest single-slot CompactPCI 

dual-core processor, dual PMC/XMC, single 

board computer. The PP 452/03x uses a popular 

mobile dual core processor from the Intel 

embedded roadmap, the 1.5 GHz or 1.06 GHz 

Intel Core 2 Duo processor. The board is ideal 

for low power intensive processing applications 

where the dual processor cores can access up to 

4 Gbytes of soldered DDR2-400 ECC SDRAM 

at up to 6.4 Gbytes/s. 

News ID 13 


congatec: COM with hypervisor software 

congatec has announced the successful testing 

of its recently launched conga-QA computer 

module with the new Hypervisor Software 2.0 

from Real-Time Systems. The Qseven embedded 

PC module conga-QA now allows any two 

operating systems to be run in parallel independently 

from one another and with real-time 

capability even where space is at a premium. 

Based on the most up-to-date Intel Atom 

Z530 processor and the Intel System Controller 

Hub US15W, the new conga-QA now also provides 

support for Intel hyper-threading and 

virtualization. 

News ID 836 

Hall-Stand 11-306/12-404 

Kontron: Box PCs in different sizes and 

configurations 

With the Concept Box, Kontron presents a new 

series of fanless compact Embedded Box PCs 

that come in different physical sizes and can be 

flexibly configured according to individual 

functional requirements. The Kontron Concept 

Box gets its high level of flexibility from the 

modular housing concept. This concept is 

based on standard housing profiles that come in 

different lengths and widths as required. Standard 

profiles are also used on the front panel 

with individual openings for dedicated I/Os as 

well as at the rear. 

News ID 874 

Hall-Stand 9-477/12-142/12-314/12-306 

Fujitsu Siemens: guaranteed security 

around the clock 

The PC control unit of NEXVS, a hybrid 

recorder for digital image recording by 

videotronic infosystems, operates with Fujitsu 

Siemens Computers’ ATX Board D2156-S21. 

The use of longlife components ensures reliable 

operations even under hardest conditions such 

as 24/7 usage or high temperatures. Additionally, 

the board’s guaranteed long-term availability 

of up to five years saves videotronic formerly 

necessary annual evaluations. 

News ID 855 



Protech Systems: Intel-ready embedded 

PC board 

Protech Systems introduced a high-speed, lowenergy-consumption 

PC board that accepts a 

full range of current Intel processors while offering 

lots of expansion options. The PSB- 

601LF, which comes with the Intel Q965 chipset 

operating at an FSB clock speed of up to 

1066MHz, has been designed to suit the needs 

of the most intensive consumer and industrial 

applications. Accordingly, Protech has configured 

the PSB-601LF as a full-size board, by 

which it can accommodate either an Intel 

Core 2 Duo, Celeron D, Pentium D or a 

Pentium 4 processor. 

News ID 956 


VIA: Pico-ITXe board with stackable I/O 

expansion 

VIA Technologies announces the VIA EPIA- 

P710 SBC based on the new Pico-ITXe specification. 

Supporting up to four customizable I/O 

expansion modules, the VIA EPIA-P710 Pico- 

ITXe enables a highly flexible and affordable 

implementation of serial connectivity options. 

Designed as the perfect baseboard, the VIA 

EPIA-P710 uses an intelligent board layout to 

allow efficient module stacking and aid heat 

dissipation. 

News ID 726 


AAEON: fanless embedded Box PC for 

extreme temperatures 

AAEON has announced the launch of a new 

BOXER series. The first debut in the BOXER extreme 

series is the AEC-7450. The embedded 

controller represents the main features of high 

speed, fanless operation in extreme hot and cold 

environments (from ‘45 C to 70C). The AEC- 

7450 is based on the high-speed Intel Pentium M 

processor (up to 1.8GHz). The system memory 

is DDR SDRAM 1GB with ECC function. 

News ID 4 


Schroff: switched-mode power supply 

unit with higher performance 

Schroff has extended the maxpowerPRO family 

of switched-mode power supplies by 

adding a PSU for CompactPCI applications 

with high power requirements to the range. 

Despite its considerably higher output capacity 

the dimensions remain unchanged (6 U/8 

HP/160 mm deep). By increasing efficiency to 

84 %, performance could be substantially enhanced. 

The new maxpowerPRO PSU now has 

an output of 500 W and features a wide input 

voltage range of between 90 VAC and 264 VAC 

to allow use worldwide in a wide spectrum of 

applications. 

News ID 916 


��... 

�� 

�� 

Meet us at our booth 

hall 12, booth 512 

PCAN PCI 

CAN interface for PCI-slots. 

Available as 1-channel, 2channel 

and optoisolated 

version. 

PCAN PC/104-Plus 

CAN interface for PC/104-Plus 

systems. Available as 

1-channel, 2-channel and 

optoisolated version. 

PCAN Explorer 4 

Universal CAN monitor, 

symbolic representation, VBS 

interface, integrated data logger, 

capability to add plug-ins 

(e. g. Instruments Panel). 

�� 

Otto-Roehm-Str. 69 

64293 Darmstadt / Germany 

Phone: +49 6151 8173-20 

Fax: +49 6151 8173-29 

info@peak-system.com



RadiSys: multi-core, turnkey platform 

for industrial 

RadiSys announces that it has developed, with 

industry consortium Embedded4You, a multicore, 

turnkey platform for industrial automation. 

The demonstration showcases the integration of 

general purpose Microsoft Windows and 

RadiSys' Microware OS-9 real time operating 

system running concurrently on a RadiSys 

COM Express module. This hybrid approach allows 

individual systems to merge onto a single 

platform to reduce hardware cost, reuse available 

software and improve system performance. 

RadiSys and Embedded 4 You are demonstrating 

the platform at the SPS/IPC/Drives exhibition. 

News ID 834 


Protech: CPU board with RMI’s AU1250 

RISC processor 

Protech has launched the ER-12LF which utilizes 

the RMI AU1250 processor. In the embedded 

computing industry, Protech has developed 

embedded CPU board for more than 

20 years. This RISC CPU module features low 

power consumption and cost-efficiency, to fulfill 

the requirements of rugged and more costfocused 

applications. The module has 5 COM 

ports, 1 USB2.0 host, 1 USB2.0 OTG, 8-port 

Digital I/O, 10/100 Ethernet, to connect and 

control peripheral devices. IDE and 1 SDIO 

slots allow. 

News ID 28 

Hall-Stand 9-429/12-565 

Advantech: COM Express module with 

Sempron/Turion dual core engine 

The new COM Express module SOM-5781 

from Advantech is ready to go. SOM-5781 has 

an onboard chip which will dynamically adjust 

the fan speed to regulate the CPU temperature. 

Because the CPU fan runs at high speed only 

when the system is under high loading, power 

consumption and noise created by the fan can 

be drastically reduced. 

News ID 808 


MEN: managed fast Ethernet switch with 

8 ports 

MEN expands its product line of robust Ethernet 

switches by another CompactPCI card. The 

F302 offers eight Fast Ethernet channels on 

RJ45, M12 or D-Sub, corresponds to the EN 

50155 railway standard and provides a number 

of options for individual port configuration via 

a special service interface at the front. MEN 

Mikro Elektronik offers a range of CompactP- 

CI and stand-alone Fast Ethernet switches for 

mobile and communication-intensive applications. 

The F302 3U CompactPCI card is the latest 

managed switch of this product line. 

News ID 918 


Acrosser: fanless and dustproof PC/104+ 

embedded computer 

Acrosser announces the release of a new 

PC/104-Plus product: AR-ES8020FL. With its 

high reliability CPU (RDC R8610 supporting 

133MHz), 64MB onboard memory, and 

128MB onboard Flash Storage, the AR- 

ES8020FL is perfect for many embedded computing 

applications. 

News ID 870 


Elma: cooling elements as front or 

rear panel 

Elma presents a solution which is based on the 

needs of 19’ technology. The cooling fin extrusion 

has a width of 42 HP and matches the 

height units. Furthermore, each and every fin is 

built upon a 1.5 HP pitch to guarantee enough 

space in between for the screw heads. Therefore, 

Exhibitors List – embedded world 2009 

/k/ Embedded Java Solutions 12-560 

3S Smart Software Solutions 11-306/10-116 

4D Engineering 11-408 

4D Systems 12-525 

AAEON Technology 9-137 

Abacus Embedded 12-236 

ABLE Design 9-235 

ABS 12-108 

AbsInt 10-412 

ACAL 9-164 

Accemic 12-508 

ACME SYSTEMS 12-525 

acontis 11-440 

Acrosser 9-267 

Actel 9-306 

Active DSP 10-239 

ACTRON 12-304 

ADDI-DATA 9-625 

Adeneo Embedded 11-318 

Adkom Elektronik 9-181 

ADLINK 9-272 

admatec 9-547 

ads-tec 9-473 

AMD 9-231 

Advanet R&D 12-116 

Advantech Europe 9-429/12-565 

Adveda 11-102 

AFT Atlas Fahrzeugtechnik 10-435 


the fins do not have to be modified for mounting, 

it is sufficient to drill the holes from the flat 

side. Hence the cooling fin extrusions can be 

used directly as front and/or rear panels in 

cabinets and sub racks. 

News ID 939 


NI: digitizers offer high-speed 

measurement for automated test 

National Instruments has introduced three 

new digitizers/PC-based oscilloscopes. The 500 

MHz NI PXI-5153 and NI PCI-5153 and the 1 

GHz NI PCI-5154 modules round out the NI 

515x series of digitizers, a complete line of highspeed 

digitizers ‘ with 300 MHz, 500 MHz and 

1 GHz bandwidths ‘ for demanding automated 

test and streaming applications where highbandwidth 

measurements are required. 

News ID 899 


MSC: secure Embedded platform for 

COM Express modules 

MSC presents Secure Embedded Platform 

based on MSC CX-MB-SP1 baseboard, specifically 

designed for applications in Gaming, Selfservice 

and Infotainment terminal industries. 

Thanks to the modularity of the baseboard and 

a wide selection of Computer-On-Module 

from MSC, designers are enabled to quickly 

build their complete embedded systems with a 

high degree of flexibility. The Secure Embedded 

Platform incorporates special security features 

for protection against unauthorized access and 

external manipulation. In addition to an integrated 

Trusted Platform Module, other features 

include: TrustedCore BIOS from Phoenix Technologies, 

battery backed security controller and 

a smart card, which can contain additional 

customer-specific security functions. 

News ID 941 

Company Hall-Stand Company Hall-Stand Company Hall-Stand 

Agilent 10-225 

Ahlers EDV Systeme 10-128 

aicas 11-109 

AIC-Xtore-PSG 9-480 

aJile systems 9-319 

All Flex Flexible Circuits 9-486 

Allegro Microsystems 12-422 

Altera 9-259 

Altium 11-236 

Altreonic 10-401 

AMBER 12-538 

AMCC 12-422/9-375 

American Arium 11-125 

Anovo 12-126

AnSem NV 12-560 

Aonix 11-119 

APAC OPTO 12-441 

Apacer 9-629 

APdate! 12-123 

Aplex 9-373 

apra-norm 12-137 

APRO 9-250 

aquintos 10-115 

Arkus 12-266 

ARM 9-273 

ARTiSAN 10-219 

ASSET InterTech 11-125 

Ass. Ghent University 12-560 

Astrodyne 12-441 

Atlantik 12-443 

Atmel 12-130/12-142 

AUO / PROMATE 12-306 

austriamicrosystems 12-142 

Avantgarde 12-322e 

AVIX-RT 11-208 

Avnet Embedded 12-236 

Avnet Memec 12-422 

Avnet 9-578 

AXIOMTEK 12-242 

Axotec 12-111 

Barco-Silex 12-560 

basysKom 11-306 

Bayer DSP 12-243 

bbv 11-123 

BCM 9-209 

Beck 9-537 

Beck IPC 9-419 

Beckhoff Automation 9-329/11-318 

BEG Bürkle 9-406 

Bicker Elektronik 12-454 

Birdstep 11-121 

blue river 10-403 

Bluegiga 12-346 

BMK GROUP 9-232 

Bolymin 12-142 

Bopla 12-455 

Brandt-Data 10-113 

BRESSNER 12-428 

Brunel Communications 11-215 

Byte Paradigm 10-525 

CAD-UL 9-501 

CalPlus 10-546 

CC&I 11-422 

CIRIC 11-406 

Cirrus Logic 12-422 

CMEL Chi Mei 12-306 

CMX Systems 10-305 

Coilcraft 12-422 

comlet 11-210 

Comp-Mall 12-349 

Concurrent 12-225 

CONEC 12-246 

congatec 12-122 

Connect 9-157 

connectBlue 9-419 

Connfly Electronic 12-558 

Contendo Systems 12-260 

Continental 11-414 

COSMIC SOFTWARE 10-216 

Coverity 11-426 

Curtiss Wright 9-283 

Dacom West 12-516 

Data I/O 10-531 

Data Modul 9-557 

Data Respons 9-457 

Datakey 9-136 

Datalight 11-125 

Daub 12-437 

DEDITEC 9-107h 

Dekimo 12-560 

Delkin Devices, 12-123 

Delta Components 12-250 

Deltatec 9-119 

demmel 9-545 

Densitron 12-441 

DENX 9-375/11-306 

Deutschmann 9-479 

Devolo 12-222 

Dexcel 9-383 

DHS ElMea 11-235 

Diamond 9-277/9-525 

Digi International 12-244 

DIGITAL-LOGIC 12-336 

Display 9-633 

dpunkt.verlag 10-519 

DSM 12-208 

DSP 11-318/12-560 

dSPACE 10-125 

DSPECIALISTS 9-519 

E.E.P.D. 9-463 

Easics NV 12-560 

EASYCODE 10-231 

Eberspächer 10-544 

EBV 9-163 

Echelon 9-182 

ECKELMANN 9-204 

eCosCentric 11-300 

eCOUNT 12-360 

EDT 12-306 

EEMBC 10-305 

EKF 12-313 

Elan 12-441 

ELBACOM 11-318 

ElectronAix 9-637 

ELECTRONIC ASSEMBLY 12-506 

Electronics Infomedia 10-448 

Elektrobit 11-414 

Elektronik i Norden 11-F224 

ELEKTRONIKLADEN 12-503 

Elma Trenew 9-261 

ELMOS 12-441 

ELPA 9-110 

ELTAN 9-255 

ELTEC 11-306 

Embedded Office 11-328 

Embedded Tools 10-239 

embedded-logic 12-434 

embeX 12-459 



EMC 9-305 

Emenda 10-114 

Emerson 9-578 

EMKO 12-259 

emlix 11-222 

EMS 12-325 

emsys 11-235 

emtrion 10-120 

EMTrust 9-463 

Enclustra 12-258 

EnOcean 12-222 

ERCO & GENER 12-270 

Eremex 11-309 

Ericsson 12-441 

esd 9-478 

ESOPE 9-107i 

Essensium 12-560 

Esterel 10-419 

ETAS 10-415 

EtherCAT 9-329 

Ethernet Powerlink 11-306 

Eurocircuits 12-255 

EUROS 11-224 

Eurostandard Press 2000 11-F223 

Eurotech 12-116 

eVision 10-540 

EVOC 9-135 

EWL 12-424 

EXAR 9-563 

Exar 12-441 

ExpertControl 10-221 

Explore 12-142 

Express Logic 11-124 

Extension Media 10-405 

exxact 12-374 

F&SElektronikSysteme 9-173 

FH Augsburg 11-203 

Farnell 12-552 

Fastwel 12-542 

FERCHAU 11-307 

Fimor 9-248 

Firecomms 12-441 

FlowCAD 11-226 

Forschung für die Zukunft 11-203 

FORTEC 9-263 

Franzis' Verlag 11-129 

Fraunhofer-Institut 12-110/11-101/12-203 

Freescale 9-335 

frenzel + berg 9-402 

FTDI 12-442 

Fujitsu 9-477/12-142/12-314/12-306 

FUTURE 9-563 

Garz & Fricke 9-529 

GCD Hard- & Software 9-379 

GE Fanuc 9-179 

Geensys 10-214 

Gennum 9-167 

Ginzinger 12-268 

Gleichmann 9-235 

GLYN 12-306 

Goetzfried 10-513 

GÖPEL electronic 10-508 

21 February 2009



GrammaTech 11-125/10-542 

GREATech 12-128 

Green Hills 10-319 

Hannusch 9-531 

HanRun 12-441 

HANSER VERLAG 10-515 

Hantronix 12-142 

Hartmann Elektronik 12-549 

Hays 10-423 

HCC-Embedded 10-305 

HCP 9-319 

Heitec 9-476 

HERMES SoftLab 11-221 

HighTec 10-407 

HILF! 9-253 

Hilscher 9-307 

Hitachi 12-306 

Hitex 10-305 

HMS 12-226 

Hochschule für Technik 11-203 

Hochschule Pforzheim 9-131 

Höft & Wessel 9-319 

HSM Zamecki 9-602 

Hüthig 10-445 

HY-LINE 9-319 

IAR Systems 10-209 

IBASE 12-417 

IBM 10-129 

IBV 11-428 

ICC Media 10-439 

ICNexus 9-180 

Icop Technology 12-540 

ICP 9-158 

ICT 12-359 

IEI 9-158 

IFTEST 9-244 

ILFA 9-381 

IMACS 11-224 

IMCOR 10-521 

IMST 9-234 

INCHRON 10-218 

INCOstartec 12-204 

Independent Electronic 12-264 

Industrial Computer Source 9-267 

Ineltek 12-142 

Infineon 12-426 

Ingenieurbüro 10-540 

InnoDisk 12-113 

Innovasic 9-281 

Inova 12-142/9-535 

INPERIO Systems 9-463 

Inspired Energy 12-105 

Insyde Software 11-125 

INSYS 12-414 

Intecs 10-511 

Intel 9-357 

INTERVALZERO 11-304 

Intrepid 10-305 

IPC2U 9-405 

ipcas 12-555 

Ircona 9-170 

Ironwood 9-305 

IS2T 11-336 

iSYSTEM 10-217 

itemis 10-529 

ITK Engineering 10-208 

iWave 12-125 

IWSFE 12-222 

IXXAT 9-279 

Janich&Klass 12-532 

Janz 9-467 

Jauch Quartz 9-257 

JET WAY 12-254 

Jungo 10-521 

K&A 10-205 

K2L 11-414 

Ka-Ro electronics 11-125/9-168 

KEIL 10-231 

Keith & Koep 11-318 

Kinsun Industries 12-441 

KIRRON 12-435 

Klar Automation 10-301 

Kleinhenz Elektronik 12-508 

Klocwork 11-F222 

Kolb Wellpappe 9-118 

König 11-225 

Kontron 11-306/12-404 

konzeptpark 11-306 

KUKA Roboter 11-440 

Kurz Industrie Elektronik 12-425 

KW-Software 11-306 

Lantronix 12-263/12-565 

Lasnier und Popa 11-F128 

Latschbacher 9-302 

Lattice 12-422/10-524 

Lauterbach 10-325 

LCD Mikroelektronik 12-303 

LDRA 10-318 

Lead 12-131 

LeCroy 10-315 

lesswire 12-403 

LIFE 9-205 

Linotype 11-305 

LinTech 9-404 

Linutronix 11-306 

LIPPERT 12-228 

Litemax 12-446 

logi.cals 12-214 

Logic Technology 11-125 

LPKF 12-518 

Luminary Micro 12-422/9-140 

Lynuxworks 11-F114 

M2M Alliance 12-322b 

macio 11-231 

Magic Power 9-107a 

Magneti Marelli Holding 11-414 

Marvell 12-422 

MAXIM 12-422 

MAZeT 9-376 

MC Technologies 12-449 

McObject 11-233 

mdex 12-322f 

MEDAV 11-111 

Meilhaus 9-350 


Memphis 12-247 

MEN Mikro Elektronik 12-545 

Mentor Graphics 11-218 

merath 12-549 

Method Park Software 11-323 

metraTec 9-107f 

MEV 12-441 

MICRIUM 11-328 

Micro Analog Systems 12-142 

Microchip 9-451 

MicroConsult 10-231 

Micronas 12-227 

MicroNova 10-221 

Microsemi 12-441 

Microsoft 11-414/11-318 

MicroSys 12-214 

Microtronics 12-322a 

MID 10-205 

Mikrap 12-421 

MikroElektronika 12-525 

MindShare 11-125 

MKS 10-311 

MontaVista 10-305 

Morex 9-116 

MosChip 12-304 

MOStron 9-268 

Moxa 12-565/12-450 

MPC Data 11-318 

MPL 9-203 

MSC 9-235 

MT Publications 10-505 

MTF 9-107d 

MTM 9-226 

M-Tronic 9-181 

Multi-Tech 12-222 

National Instruments 10-221 

NAVIGON 11-414 

NEC 12-422/9-447 

NetModule 12-524 

New Japan Radio 12-441 

Nokia 11-110 

Nolam 9-184 

Nordic 12-422 

Noser Engineering 11-F119 

Nu Horizons 9-167 

Nuance Aachen 11-414 

NXP 12-218 

Objective Software 11-302 

OBP 10-504 

OLIMEX 12-503 

oose 10-519 

Open Source 11-306 

OpenSystems 11-410 

Optimem 12-144 

OSC 10-118 

Overseas Marketing 10-344 

Panasonic 9-319 

Parasoft 11-209 

PC/104 Consortium 12-135 

PEAK-System 12-512 

Pengutronix 11-306 

Penton Media 11-F123

Perforce 10-343 

Peschges Variometer 9-461 

Phytec 11-306/12-554 

Picoway Technology 11-F103 

PikeTec 10-516 

PLC 2 9-167 

PLDA 12-559 

Plextek 11-318 

pls 10-215 

PLUG-IN 9-354 

PMC 12-422 

Pohl Electronic 9-647 

POLARION 10-523 

port 12-224 

Portwell 9-278 

Power Integrations 12-441 

powerBridge 12-550 

PRAGMADEV 10-239 

Premier 9-643 

Premo 12-142 

ProAnt 12-458 

Proemion 12-141 

profichip 9-219 

Programming Research 10-119 

Projektron 10-400 

ProMik 9-308 

Protech 9-561 

pure-systems 10-124 

QA Systems 10-119 

QNX 11-324 

QTronic 10-305 

Quad 12-424 

Quadros 11-431 

QualiSystems 10-510 

QuartzCom 12-531 

Radisys 9-229 

RAISONANCE 10-128 

Ralink 12-441 

Rapita 10-421 

RDC 12-304 

RealiteQ 9-419 

Realtek 12-441 

Real-Time Systems 11-F110 

Red Rocket 12-441 

Reed 11-F108 

Reikotech 9-125 

Renesas 12-306/9-235 

Rittal 12-430 

RM Components 12-251 

RM Michaelides 12-141 

Ronetix 11-224 

RoseTechnology 9-567 

RS Components 12-354 

RUTRONIK 9-435 

Samtec 12-568 

Sasco 12-362 

Satyam 11-115 

Scantec 12-221 

SCATTERWEB 9-107k 

Schleißheimer 10-446 

Schmid Engineering 10-221 

Schroff 12-318 

Sdelovaci 10-507 

SE 9-282 

SECO 12-367 

SEGGER 10-310 

semica.de 9-667 

Semtech 12-422 

senTec 12-503 

sepp.med 10-437 

SEVENSTAX 12-418 

Siemens 11-108/11-318 

SIEPA 12-468 

Signum Systems 10-239 

Silex Technology 9-319 

SILICA 12-330/12-230 

Silica 11-318 

Silicann 9-107c 

Silicon Laboratories 12-136/12-422 

Silicon Microstructures 12-441 

Silicon Storage Technology 12-355 

Simple Solutions 12-522 

Simplify Technologies 9-631 

Sitek 9-419 

SMSC 12-422 

SMT & HYBRID 9-380 

SoftComponents 9-107b 

solvimus 12-357 

Sontheim 12-117 

SOQUS 11-438 

SORCUS 9-230 

Sparklan Communications 12-441 

Sparxsystems 10-404 

Spectra 9-318 

SPHINX 12-565 

spo-comm 9-236 

SPOERLE 12-362 

SSV Software Systems 12-544 

STEC 12-441 

Steinbeis-Transferzentrum 12-322h 

STMicroelectronics 12-326 

Stollmann E+V 12-547 

Strampe 12-541 

STZ 9-131 

Summit Microelectronics 12-441 

Swissbit 9-435 

SYMTAVISION 10-412 

SYS TEC 10-211 

SYSGO 11-128 

system elektronik 9-551 

System Industrie Electronic 9-217 

SystemBase 9-172 

SYSTERRA COMPUTER 9-179 

TABO 9-107j 

TARA Systems 10-512 

TASKING 10-305 

taskit 12-402 

TechConnect 10-538 

Techinsights 10-444 

TU Dresden 11-203 

TUt München 9-108 

tecnotron elektronik 9-129 

TecSys 9-128 

Tektronix 10-411 



TELCONA 12-365 

Telenor 12-322d 

TenAsys 11-315 

Texas Instruments 12-436 

The IET 10-450 

The MathWorks 10-335 

Thesycon 12-418 

Tibbo Technology 12-441 

Tieto 10-204 

TIGAL 12-525 

TIGRIS 12-107 

TILCON 10-239 

TOPSCCC 12-456 

Toradex 9-130/12-133 

Toshiba 12-460/12-306 

Total Phase 10-222 

TPO Displays 12-142 

TQ-Components 12-342 

Trenz Electronic 12-536 

Tri-M Engineering 9-673 

TRINAMIC 12-441/9-107g 

TriQuint 12-422 

TRL 12-322c 

TTTech 10-317 

Tundra 12-422 

ULTRATRONIK 12-121 

Unicoi Systems 11-422 

Unitronic 12-222 

Universität Leipzig 11-203 

UPEK 12-422 

VAST 10-518 

Vector 10-339/10-441 

Verifysoft Technology 11-112 

VIA 12-350 

Virtium Technology 9-218 

VISION SYSTEMS 9-132 

Visual 10-540 

Visure Solutions 10-119 

Vliesstoff Kasper 9-531 

Vogel Business Media 10-206 

voice INTER connect 11-418 

WAGO Kontakttechnik 9-207 

Wavecom 12-306/12-422/12-222 

WEKA-Verlag 11-129 

Willert 10-231 

Wind River 11-118/10-239 

WITTENSTEIN 11-136 

Wordsworth Technology 9-202 

WÜRTH Elektronik 12-546 

Xentech 9-144 

XILINX 12-515 

XiSys 12-214 

X-SPEX 12-107 

Yellowstone 9-105 

YOKOGAWA 10-416 

Zarlink 12-422 

ZENSYS 12-441 

ZEROPLUS 10-509 

ZiLOG 12-422 

Zitzmann 11-218 

Zühlke Engineering 10-429 

23 February 2009

AVIONICS 


communications in UAV systems 

By Bert Farabaugh, Real-Time Innovations 

An increasingly wide range of 

unmanned air vehicles (UAVs) 

are posing unique design and 

development requirements 

particularly regarding the data 

link. Many defense 

application developers have 

turned to the open data 

distribution service (DDS) 

standard to provide the 

necessary messaging 

framework. 

■ The unique design and development requirements 

of the wide range and types of unmanned 

air vehicles (UAVs) are pushing the capability 

boundaries of available technologies in many 

areas, none more than the demands of the data 

link. Developers invest heavily in the data link as 

a defining element of their competitive value in 

their UAV system, because it has direct impact 

on the range, flight time and sensor feedback capabilities 

of the UAV. Delivering the benefits of 

that investment to the rest of the UAV system is 

the function of the messaging system that connects 

the data link to the ground station and 

UAV application environments. 

In a system design, the physical transport and 

signalling layer of the data link may or may not 

manage all the challenging issues of communication 

over the air in order to present a clean 

message to the application layer. There will be 

critical power/weight/performance trade-off 

decisions to make in the UAV data link that may 

not apply in the ground station. For software 

system architects, the design and development 

of a messaging framework that will support 

these complex data link design trade-offs and 

additionally provide an application integration 

capability between the ground station and the 

UAV is a major investment. This can become 

one of the most challenging tasks in UAV system 

design and development. As a data transport, 

the radio/satellite-based data link has to be 

integrated into two very different environments: 

the deeply embedded, power and weight 

constrained UAV, where every bit of data sent 

directly detracts from the power needed to enable 

extended flight time objectives, and the 

ground station, which is usually tethered and 

operating in an environment somewhat similar 

to a ruggedized workstation. The data-link signal 

has to contend with a wide range of com- 

Data Link 

Communication 

Issues 


design and 

implementation 

design 

constraints 


Intermittent or 

lossy data links - 

Packet loss is 

common and 

connection loss is 

not uncommon 

Integrated 

into a power 

and weight 

constrained 

UAV 

Integrated 

into a tethered 

ruggedized 

workstation 

environment 

Maximising 

bandwidth 

utilisation - 

Ensuring that the 

packet messaging 

overhead is 

minimized 

Table 1. Communications problems of the data link signal 

Figure 1. RTI real-time middleware 

is being used in the 

General Atomics Aeronautical 

Systems, Inc. (GA-ASI), 

Advanced Cockpit Ground 

Control Station (GCS). 

munication problems (see examples outlined in 

table 1), such as intermittent or lossy connectivity, 

and one-way receive communications 

when the UAV must fly in a silent non-transmitting 

mode. This type of connection could 

potentially require a complex and large packet 

structure overhead to deal with the underlying 

physical transport issues, which is directly 

counter to some of the UAV design constraints 

Area of 

Responsibility 

Shift - 

seamless 

transfer of 

control of UAV 

from ground 

station to 

ground station 

Dealing with multiple 

data-paths in 

the data link - 

Managing out-oforder 

packets that 

result from radio 

signal reflections 

Beyond 

Line of 

Sight and/or 

Line of Sight 

radio 

communication 

Unidirectional 

communication - 

Receive only mode 

in the UAV can 

save power, or be 

used to assist in 

UAV cloaking 

Re-configurability 

– support 

re-configurable 

payloads, which 

places additional 

and variable 

demands on the 

data link for 

sensor feedback 

or command 

mechanisms 

Table 2. Connection potentially requires a complex and large packet structure overhead to deal 

with the underlying physical transport issues.

(table 2) such as operation within a power-constrained 

system. Every transmitted bit is sacred. 

It is therefore very important that access to the 

transport layer of the messaging is facilitated so 

this can be optimised for the needs of the particular 

application. It is just as important that 

the messaging layer itself is constructed to 

specifically meet these design objectives, putting 

minimum demands on the link in order to 

meet the specific send/receive requirements for 

messages across that link type. 

In addition to these issues, there are in fact several 

link types, each with their own unique set 

of performance, integrity and reliability requirements, 

some of which are detailed in table 

3. So the messaging system built to run over a 

specific link type has to be tailored to the specific 

demands of that link in the most optimal 

way possible. These requirements maybe aggregated 

into fewer actual physical transports as 

long as that transport can meet all the requirements 

listed for each service type. With this 

wide range of requirements and constraints, it 

is tempting to believe that only a set of very 

specifically developed and separately optimized 

messaging systems, one for each type of 

connection, is required. However this is not the 

case. The messaging model for the data link 

should encompass a number of key characteristics. 

It must be real-time, capable of supporting 

high-speed data transfer while also providing 

high-integrity messaging. At the same time 

it has to deliver high availability (or re-routing) 

and recovery over a physical link that is not itself 

guaranteed. In addition, for emerging civilian 

requirements the code base should be 

ready for safety certification too. 

Many UAV and other defense application developers 

have turned to the open data distribution 

service (DDS) standard (managed by 

the Object Management Group, OMG), to 

provide a messaging framework that can successfully 

deliver the real-time capabilities demanded 

by the data link to the mission critical 

application environment. Existing applications 

include: Insitu Scan Eagle UAV, General Atomics 

Predator UAV, US AirForce / Navy common 

link integration processing, US Navy Aegis 

weapon system and the F-35 Joint Strike Fighter 

electronic warfare system. DDS specifies a 

publish-subscribe messaging model. The critical 

and differential DDS messaging approach 

is to make this model executable in a real-time 

system context. It does this by providing a set of 

Quality of Service (QoS) parameters that are 

negotiated and agreed upon between every 

publisher and subscriber. There are a large 

number of configurable QoS parameters to 

meet a wide range of real-time message requirements, 

but for our purposes we will focus 

on just a few of them executed in a publish/subscribe 

model, to show how DDS has been used 

by companies like Insitu and General Atomics 

in the ScanEagle and Predator/Reaper unmanned 

aircraft systems. The most basic observation 

about messaging between a UAV 

and a ground station is that one system environment 

is deeply embedded (UAV) and the 

other tends to be more of a ruggedized workstation 

system (ground station). This will mean 

that the implementation of the messaging 

model may vary while the functional task remains 

the same. For example; how much signal 

reflection management is done at the signalling 

level? In the ground station you can 

probably afford to integrate a very high performance 

reflection removal DSP and associ- 

AVIONICS 

ated algorithm into the system design, so that 

all received messages are cleanly presented to 

the application layer. However in the UAV you 

maynothavethepowerorweightbudgetfor 

that DSP, so this would now put a greater burden 

on the messaging framework to recognise 

and discard out-of-order packets and duplicated 

data. Perhaps this is less efficient in data 

throughput, but that would be the trade-off 

decision with the power and weight budget. 

The most fundamental issue to contend with 

on the data-link between the UAV and ground 

station is that it is lossy and transient. The 

effects of the nature of the link should not be an 

25 February 2009

AVIONICS 

Figure 2. The General Atomics Aeronautical Systems MQ-1 Predator UAV 

issue an application (like the flight computer) 

should have to deal with. So the messaging layer 

should hide and resolve problems such as lost 

or out-of-order packets from the application 

layer as much as possible. Let us take the example 

of commands sent from the ground station 

to the UAV flight computer, which will 

usually require some form of correlated command 

result feedback to the ground station. 

With DDS we would define the publisher/subscriber 

relationship as RELIABLE. This tells 

DDS to select a reliable packet model (one of 

several real-time options) on top of the transport 

to deal with the issue of lost packets. However 

the application may still want a degree of 

control over the link and how to respond 

should the link fail. So you could set a QoS of 

DURABILITY = TRANSIENT_ LOCAL with a 

Data Link Type 


Utilization 

Performance 

and Integrity 

requirements 

High Reliability 


(HRDL) 

– 

primarily responsible 

for 

C2 data and 

status data such 

as position and 

altitude 

Moderate 

throughput 

with high availability 

and integrityrequirements 

High Capacity 


(HCDL) 

– 

sensor data 

transfer 

High throughput 

(possibly 

streaming) requirements 

for 

the downlink 

further qualifier N for the amount of historic 

messages to keep. This would direct DDS to 

keep the last N messages stored locally in 

memory; this message data is then available for 

an application re-synch after re-establishment 

of lost connection for example. The flight 

computer application may also want to define 

what time-duration of link loss constitutes a 

link failure from its point of view. So you would 

set a QoS as LIVELINESS lease duration > 

common/acceptable link down-time (for 

example 5 seconds). 

These QoS configuration options demonstrate 

the flexibility of DDS to be adapted to some of 

the real-time message issues in a UAV data link. 

It also shows how DDS allows an application 

the ability to define its localised real-time con- 

Beyond Line of 

Sight Data Link 

(BLOS DL) 

data exchange via 

a relay platform 

such as a satellite, 

high altitude 

platformora 

relay UAV 

Aggregate performancerequirements 

of 

HRDL and 

HCDL 

Table 3. Examples of link types, each with their own unique set of performance, integrity and 

reliability requirements 


Back Up Data 

Link (BUDL) 

C2 and status 

data transfer in 

emergency 

Low throughput 

with very high 

reliability and integrity 

straints in terms familiar to the application developer. 

DDS automatically executes against 

those constraints and validates that both the 

subscriber and publisher, and by implication 

the physical transport, can meet those requirements, 

reporting back to the application when 

those real-time communication objectives are 

not met. A complete messaging framework will 

use the data-link as just one (albeit critical) 

transport among many across a UAV system design.Itneedsalsotoprovideadistributedmessaging 

capability across the ground station and 

internal to the UAV. For example in the UAV 

there will probably be the data link controller, 

flight computer (supported by an RTOS), an 

application platform and potentially multiple 

(smart) sensor nodes connected via a variety of 

bus and network transports. In the ground-station 

you will have a data link controller, the 

general purpose ground station running Linux 

or Windows, a data visualization/display computer, 

and probably an interface to a broader 

net-centric environment into which the UAV 

system is integrated. So a complete system-wide 

messaging framework needs to support both 

the highly specialised requirements of the 

point- to-point data link(s), and to provide a 

distribution capability for data and information 

to flow in real-time across the entire UAV system 

and out to the net-centric environment. 

The publish/subscribe model itself is wellsuited 

to both the distributed system messaging 

requirements and the data-link point-to-point 

requirements. For example, publish/subscribe 

helps deal with one of the fundamental challenges 

of UAV system design, the transfer of 

control of the UAV between ground stations. 

With DDS a UAV application subscriber can listen 

to multiple ground station command publishers. 

Every publisher can be assigned a 

strength. If a command subscriber in the UAV 

receives messages from several ground station 

publishers the application will only receive the 

command from the strongest publisher. This 

automatically deals with the command receipt 

overlap time period. To explicitly manage 

transfer of control of a UAV from one ground 

station to another, the currently highest 

strength ground station command publisher 

could await notification from the UAV that it 

was receiving a satisfactory validated set of 

commands from another command publisher. 

The ground station that needs to take control 

can be authorised to increase its publishing 

strength, then the current ground station command 

publisher would automatically be superseded 

by commands from the new higher 

strength publisher. Such a design approach may 

be useful in swarm management too, where a 

distributed messaging model would be needed 

with a flight leader that may change over time. 

This technique would also be useful in the event 

that the high reliability data link fails and C2

messaging needs to fall back on the back-up 

data link (table 3, link types). The inference here 

is that the message system would automatically 

fall back to using a different physical transport 

(data link) in such an eventuality. This is 

exactly how DDS can be set up. The messaging 

throughput over that link may be less (a lower 

speed connection) but the DDS messaging layer 

can be set up to compensate for this too. For example 

it may be set up to only re-route only a 

specific sub-set of the critical C2 commands 

needed when using this type of connection. 

As already mentioned, power conservation is 

paramount within the UAV system, and minimizing 

the amount of data transmission off the 

vehicle can greatly help reduce power consumption. 

DDS provides additional data filtering 

capabilities that can be enacted upon by 

the publishing application. The UAV can set up 

a filter on a given data stream for the instances 

of data that satisfy its current operational constraints, 

DDS will then eliminate any unneeded 

data instances from being transmitted on the 

data link. The same holds true for the opposite 

direction, the ground station can specify a filter 

on data going to the UAV and have that data 

filtered before the data is sent across the data 

link. These filters are dynamically changeable to 

accommodate changes in mission operations. 

The end result is a dynamically self optimizing 

data link that only sends relevant data to/from 

the UAV, therefore reducing unnecessary power 

consumption. 

The benefit of DDS is that all the complexity of 

who is sending data and over which transport 

now becomes transparent to the application, effectively 

decoupling the two, which makes the 

application code far more portable and re-useable. 

If needed, the management of what entity 

is publishing data and the selection of what 

transport is being used for specific data streams 

is accessible for applications to create intelligent 

■ Mercury: multi-GPU development 

platform 

Mercury has unveiled the GPU-based Sensor 

Stream Computing Platform, a programmable 

floating-point graphics-rendering engines primarily 

used in personal computers, workstations, and 

gaming consoles. For traditional signal processing 

algorithms like the FFT, they provide unprecedented 

performance, particularly performance 

per watt. With the Mercury Sensor Stream Computing 

Platform, embedded stream computing 

customers can benchmark and evaluate application 

performance in their choice of GPU environments, 

and then migrate to a larger deployed solution. 

News ID 790 

Product News 

AVIONICS 

communication architectures, that reconfigure 

themselves based on actual mission scenarios. 

While many UAV system designs were initially 

targeted at military applications, there has 

been a rapid move to take these UAV system 

platforms and apply them to civil applications. 

Additionally military UAV utility (range and 

flight duration) is now such that UAVs need to 

fly through civil airspace. In either case the UAV 

has to be developed to standards that can be 

safety certified such as ED-12B and DO-178B 

(once these standards have evolved to address 

the specific unique issues of UAVs). The sup- 

Interested in more information 

about real-time communications? 

Visit our specific website with links to: 

�"Integrating Security into Real-time 

Distributed Systems" 

Paper to be presented on March 3 at Embedded 

World, Nuremberg, Germany 

�"UAV Data Link Design for Dependable 

Real-time Communications" 

Paper to be presented on March 12 at 

AVIONICS, Amsterdam, Netherlands 

�"An Open Integration Framework for a reconfigurable 

Multi-mission UAS Platform" 

PapertobepresentedonApril20atAUVSI, 

La Spezia, Italy 

Real-time Middleware for UnmannedVehicles 

www.embedded-control-europe.com/knowhow?kid=324 

pliers of DDS implementations to this UAV 

sector are responding to this requirement, by 

developing certifiable versions of the DDS 

messaging software system. ■ 

■ Mercury upgrades PowerPC 

multicomputer series 

Mercury Computer Systems announces the 

availability of the RACE ++ Series PowerPC 

7448 Multicomputer, a drop-in product upgrade 

that adds significant improvements in 

processor speed, L2 cache size, and application 

performance over previous RACE++ products 

built on the PowerPC 7447A processor. The 

PowerPC 7448 comprises a Mercury MCJ6 

motherboard and two faster daughtercards 

that interface directly with the RACE++ 

switch fabric. 

News ID 907 


P O W E R 

Innovation In Motion. 

cwcembedded.com 

sales@cwcembedded.com 

O N D I S P L A Y 

Today’s high performance rugged aerospace 

and defense applications demand previously 

unreachable levels of embedded processing and 

graphics display capabilities. Curtiss-Wright 

is a step ahead with their next generation of 

VPX single board computers and XMC graphics 

display/video capture solutions. The VPX6- 

185 SBC is a Freescale Power Architecture 

MPC8641D Single Board Computer delivering 

high bandwidth and performance along with 

Serial RapidIO and PCI Express ® switched 

fabrics. The VPX6-185 is an ideal host for the 

XMC-710 graphics controller card, powered 

by a NVIDIA ® G73M graphics engine, with 

dedicated video memory and Curtiss-Wright 

graphics IP. 

VPX6-185 

XMC-710

AVIONICS 

Implications of adopting aerospace 

development and verification standards 

By Brian Hooper and Bill St Clair, LDRA 

■ An ever-increasing reliance on software control 

has meant that many companies from nonaerospace 

business sectors, such as automotive, 

nuclear power, MRI scanners, financial systems, 

that do not have a traditional requirement for 

sophisticated software development processes, 

now find themselves compelled to undertake 

safety-critical and safety-related analysis and 

testing. With the need for increased software 

quality across different industries, a tendency 

has emerged for companies to look outside 

their own market sector for best-practice approaches, 

techniques or standards. Examples of 

such industry crossover have been seen in the 

automotive and avionics industries, with the 

adoption of elements of the DO-178B standard 

by the former and a similar adoption of the 

Motor Industry Software Reliability Association 

(MISRA) standards by the latter. 

In adopting out-of-sector quality and testing 

standards, new and unfamiliar development 

and testing techniques need to be implemented: 

conforming to a set of coding standards, 

such as MISRA-C or Joint Strike Fighter Air Vehicle 

Coding Standards (JSF++ AV). Along with 

this are required an automated checking tool, 

formal unit testing plus informal debugging to 

demonstrate that requirements are satisfied as 

they are incrementally implemented, code coverage 

that validates the effectiveness of testing 

and isolates non-executable code, and code cov- 

February 2009 

Non-aerospace industries are 

increasingly obliged to 

undertake safety-critical and 

safety-related analysis and 

testing. Companies in these 

industries are consequently 

tending to look outside their 

own market sector for 

best-practice approaches, 

techniques, and standards. 

This article examines 

the challenges. 

erage reports that trace all aspects of each line 

of source code for safety-critical components. 

Let’s look at each technique in detail to understand 

the specific challenges involved and learn 

ways to overcome them. 

Software in airborne systems and equipment in 

the early 1980s resulted in a need for industryaccepted 

guidelines for satisfying airworthiness 

requirements. DO-178, Software Considerations 

in Airborne Systems and Equipment 

Certification, in its revised version DO-178B 

became the defining standard for aerospace systems 

and software. DO-178B is primarily a 

process-oriented document in which objectives 

are defined and a means of satisfying these objectives 

are described. Failure conditions associated 

with the system and its software components 

undergo system safety assessment according 

to the famous A-E categories, which determine 

the level of effort required to show 

compliance with certification requirements. 

Similarly, in 1998 MISRA published their C 

standard to promote the use of safe C in the UK 

automotive industry. MISRA promotes the 

safest possible use of the language by encouraging 

good programming practice, focusing on 

coding rules, complexity measurement and 

code coverage, and ensuring well designed 

and tested code. Lockheed Martin built on the 

MISRA-C guidelines by adding a set of rules for 

28 

C++ language features (e.g., templates, inheritance) 

to create the JSF++ AV coding standard. 

Its adoption ensures that all software follows a 

consistent style, is portable to other architectures, 

free from common errors, and easily understandable 

and maintainable. To provide a 

general C++ guideline, MISRA released a C++ 

standard in June 2008. 

To conform to these standards via a traditional, 

manual peer review process would be tedious, 

time-consuming and without any guarantee 

of completeness or ability to demonstrate 

to a certification authority that source code is 

100% conformant. Tools automate the code review 

process by developing a fast, repeatable 

process which can deliver useful and usable 

quality reports. The MISRA standards, when 

used within the wider process framework of 

DO-178B, provide an extended model that addresses 

both quality and reliability. Projects that 

have adopted this approach find real cost and 

reliability gains that benefit non-aerospace industries 

as the need for quality increases. 

Software teams in any industry never deliver a 

component without testing, but the testing may 

be dubious. As the final phase in software development, 

testing is squeezed as earlier phases 

overrun and delivery dates must be met. Even 

with time, the typical approach is to use functional 

testing to demonstrate the capability of

Objective Applicability by SW Level 

Modified 

condition/decision 

coverage is required 

Decision coverage is required 

Statement coverage is required 

the software to meet its requirements. This style 

of testing is usually performed at the system 

and/or subsystem levels; it is highly procedural, 

consisting of hundreds of steps, and is part 

of a top-down process of system validation. 

Functional testing is only as good as the requirements 

against which the tests are developed. 

The Standish Group Chaos Report reveals 

that only 35% (up from 16.2% in 1994) of 

software projects are completed on time, on 

budget and to user requirements. Functional 

testing, which requires that the system (or subsystem) 

under test must be coded and functional 

before testing can begin, does not address 

meeting requirements which may be missed 

due to complexity, ambiguity and imprecise 

A B C D 

✓ 

WELCOME 

✓ ✓ 

✓ ✓ ✓ 

definition or scope creep. 

DISCOVER THE FUTURE OF 

THE AVIONICS INDUSTRY 

Many aerospace companies now employ iterative 

development processes in which they focus 

on subsets of a modular system. This technique, 

typically called unit testing, is a bottom-up 

process that focuses on system internals, such as 

classes and individual functions. Unit testing facilitates 

early stage, prototype development and 

covers the paths and branches in the software 

that may be unpredictable or impractical to exercise 

from a functional testing perspective (e.g., 

error handlers). Traditional manual unit testing 

processes require high skill levels and involve 

considerable, additional overhead. Automation 

of these processes via tools standardizes the 

TEST LEARN SOURCE NETWORK 

REGISTER ONLINE TODAY WWW.AVIONICS-EVENT.COM 

85 leading avionics suppliers in one dedicated exhibition 

10+ FREE presentations from the likes of Infotech Enterprises, 

Elbit Systems, Esterel, Tyco Electronics, Saab Avitronics… 

High Level 2 day conference – with speakers from NASA, 

Airbus, Boeing, Eurocontrol, Rockwell Collins, Lufthansa, 

Horizon Air, Green Hills, Wind River etc 

AVIONICS 

techniques while leaving room for intuition - 

benefits that increased efficiency and reduce 

costs. Automation facilitates development of repeatable 

processes and standardized testing 

practices. Tools also capture and store complete 

test information in a configuration management 

system with the corresponding source 

code, which can be retrieved and imported later 

for regression testing. 

Functional and unit testing proves that software 

satisfies requirements and errors have been removed. 

To determine how effective testing effort 

has been, you need coverage analysis applied 

in tandem with a set of test cases that exercise 

requirements and highlight which sections 

of code have and have not been executed. 

The identification of non-executed code pinpoints 

a number of potential shortcomings: errors 

in the test cases, imprecise or inadequate 

requirements, inadequate requirement testing, 

and dead code, i.e., code impossible to execute. 

Code coverage has several levels of precision 

with, at a minimum for DO-178B, Level C, it 

showing that test cases executed a line of 

source code at least once. Greater precision is 

required as the safety level increases. The accompanying 

table illustrates the amount of 

structural coverage required per safety level: 

11th–12th March 2009, Passenger Terminal Amsterdam (city centre), 

Amsterdam, Holland www.avionics-event.com 

REGISTER BEFORE 1 MARCH & SAVE 15% 

9 Workshops – with speakers from Wind River, Green Hills, 

Naverus, Rapita Systems, General Electric, AdaCore, RTI, Frank 

Alexander & Associates and CMC Electronics 

3 x ½ day MasterClasses from Presagis, Verocel, Wind River, GE 

Fanuc, Seaweed, Adacore, Boeing 

NETWORKING RECEPTION on the 11th March 

Owned and Produced By:

AVIONICS 

Figure 1. Example of unit testing 

Traditional manual unit testing processes require 

high skill levels and involve considerable 

additional overhead. Automation of these 

processes via tools standardizes the techniques 

while leaving room for intuition - benefits that 

increased efficiency and reduce costs. Automation 

facilitates development of repeatable 

processes and standardized testing practices. 

Tools also capture and store complete test information 

in a configuration management system 

with the corresponding source code, which 

can be retrieved and imported later for regression 

testing. Statement coverage may be sufficient 

to identify missing test cases as illustrated 

in the following code snippet: 

if (reading > 10.0) { 

led_mode = 4; 

} 

else { 

led_mode = 5; 

} 

With red signifying covered code, it is clear that 

a new test case is needed to exercise the situation 

where “reading” is not greater than 10. 

Statement coverage is sufficient to identify 

dead code here: 


led_mode = 4; 

} 

else { 

led_mode = 5; 

} 

if (led_mode == 8) { 

update_panel(); 

} 

No matter how many additional test cases are 

created, the call to update_panel can never be 

reached.Therootoftheproblemmaybeadesign 

error in another part of the code, or the 

code is not traceable to a requirement. Typically, 

due to if-then branches and loops, there are several 

routes through a software component 

and, above safety level C, each route must be exercised 

and reported as covered. This is known 

as decision coverage and may be illustrated by 

the following code snippet: 

led_mode = 0; 


led_mode = 4; 

} 

update_panel(); 

The report shows that we have exercised the 

code with values of reading up to 10.0 but not 

above. Statement coverage would highlight 

this too, of course, but will not show the converse, 

i.e., values of reading above 10.0 but not 

below. And, we need to be sure that update_panel 

has been called with led_mode set 

to 4 and when left with its initial value of 0. The 

highest level of coverage precision is modified 

condition/decision coverage (MC/DC). This is 

reserved for software components assessed at 

safety level A and places the component under 

exhaustive testing to prove that each decision 

tries every possible outcome, each condition in 

a decision takes on every possible outcome, and 

each condition in a decision is shown to independently 

affect the outcome of the decision. In 

simple terms, we are concerned with the 

www.embedded-control-europe-com 


permutations of a compound condition as 

illustrated in this code snippet: 

if (reading > 10.0 or submode == 3) { 

led_mode = 4; 

} 

else 

{ ... 

The coverage report needs to confirm that we 

have exercised the code where reading is both 

above 10.0 and below in combination with submode 

being 3 and some other value, i.e., 4 permutations. 

Source code verification gauges the effectiveness 

of testing, whether proving all requirements 

are satisfied or by uncovering problems. 

This task cannot be undertaken manually. 

Thankfully, coverage analysis is highly automated 

in most test tools, offering virtually transparent 

usage. The tools increase the quality of code 

and integrate the overall test process, reducing application 

failure rates and maintenance costs. 

Object-code verification focuses on how much 

the control flow structure of the compiler-generated 

object code differs from its application 

source code. Given that traditional structural 

coverage is applied at the source code, but the 

object code executes on the processor, differences 

in control flow structure between the two 

can result in significant gaps. These flow graphs 

are generated from the same procedure, but the 

Object-box Mode graph on the left shows a 

branch which doesn’t appear in the right side 

Source Mode graph. Visible, easy-to-use reports 

help engineers quickly build test cases to 

achieve 100% unit coverage. Without such reports, 

the effort required to identify each path 

through the object code would result in longer 

timescales and higher cost. 

Software components in aerospace systems 

with a DO-178B Level A must be object code 

verified. While this is the toughest testing discipline, 

it now comes increasingly into consideration 

as more safety-critical components are 

deployed in automobiles, medical equipment 

and transport control systems. Similarly, critical 

components in telecom and financial systems 

have increased quality requirements due 

to the high monetary cost of failure. Although 

safety-critical components represent a subset of 

the whole application, object code level requires 

considerable resources in terms of time and 

money. Deploying automated, compiler-independent 

processes helps reduce overall development 

costs by several factors and delivers 

high-quality software components with minimal 

chance for failure. ■


hardware and software 

By Ulrich Kleber, Service Availability Forum 

The new platform 

management service added 

by the Service Availability 

Forum provides the software 

with easy access to relevant 

state changes of hardware 

entities. This article explains 

its functioning with reference 

to a typical AdvancedTCA 

shelf with a number of 

blades represented in the 

information model. 

■ The Service Availability Forum (SA Forum) 

has published a new release of specifications, 

adding a new service called platform management 

to the application interface specification 

(AIS). This new service addresses the interaction 

between hardware events, which can be accessed 

by the hardware platform interface 

(HPI), and the software entities. The HPI provides 

a view of the a system, based on discovery 

of the hardware that is present, and reports 

changes in the hardware by events, which must 

be analyzed by software before appropriate 

actions can be taken. 

Implementations of the AIS services as well as 

applications with direct hardware dependencies 

want such actions to happen automatically to 

provide high availability. Therefore it is necessary 

to provide easy access to relevant state 

changes of hardware entities to such software. 

This is where the new platform management 

(PLM) provides its service. Additionally, it 

covers the virtualization and operating system 

layers with similar interfaces, relevant for middleware 

and applications in a high-availability 

system. 

The PLM service defines an information model 

that represents the configured hardware entities. 

This model is used by a developer to create object 

instances for every piece of hardware that 

the software needs to provide its service. These 

objects store the necessary data to allow hardware 

type checking and to indicate the relationship 

between hardware entities. As an example, 

a mezzanine module sitting on a compute 

blade or carrier board is represented by an 

object in a containment tree, the carrier board 

being its parent. When the carrier board is 

physically removed from the system, all contained 

entities are affected. The PLM service 

now can reflect this in the information model 

by changing the states of all children of the 

removed board. 

The following example (figure 1) shows a 

typical AdvancedTCA shelf with a number of 

blades represented in the information model. 

The example shelf contains double switches and 

six computing blades, some providing I/O interfaces 

on AMC cards and some providing a 

disk. The entities shown in figure 2 can be 

represented in a hierarchical tree of managed 

objects as follows. The relationships of the entities 

in the tree reflect the physical dependencies 

and locations. The shelf object is parent to 

all the blades (including switch blades) and also 

to the field-replaceable units of the infrastructure, 

like shelf managers, fans and power entry 

modules. Blades in the same way parent the 

sub-modules, in this case AMC-I/O cards and 

disks. This hierarchical tree modeling allows 

PLM to provide appropriate state management 

for these objects. State changes of a blade also 

COMMUNICATIONS 

Figure 1. A typical AdvancedTCA shelf with 

a number of blades represented in the 

information model 

affect its sub-modules: a disk in this configuration 

cannot be accessed if its carrier CPU 

blade is faulty; and all AMC cards get extracted 

if their carrier blade is extracted. PLM provides 

its users with callbacks to inform them 

about important hardware events on the objects 

in which they are interested. 

Assume that in the previous example an application 

running on one of the diskless blades 

stores its data on one of the disks located on another 

blade. It may happen that the operator 

wants to exchange the disk blade and opens the 

latches. In that case PLM is informed by HPI 

hot swap management and will notify users that 

are subscribed for this entity that an extraction 

request is pending. The application can now 

safely close all files and do the necessary replication, 

before allowing the blade to be extracted. 

The application could also directly subscribe 

for HPI hot swap events, but in case there are 

multiple users of that AMC disk, a central service 

needs to decide when it is safe to extract the 

disk. PLM provides this service, being the 

owner of the managed object representing the 

disk. It acts as an HPI user that does all the steps 

in HPI that are necessary for proper hot swap 

management, and provides a standardized 

interface to all affected software services. Additionally 

to this representation of hardware 

elements, PLM similarly introduces object 

classes to model virtualized architectures, where 

31 February 2009


Figure 2. The entities can be represented in a hierarchical tree of managed objects. 

Figure 3. Virtual machine monitor and a number of virtual machines managed by each monitor 

reflected in the information model. 

virtualization facilities (VF, also called hypervisor 

or virtual machine monitor) are able to 

run multiple operating systems on virtual machines. 

The PLM information model represents 

the containment hierarchy of virtualization facilities 

and virtual machines with so-called execution 

environment objects. These objects 

allow the information model to reflect the dependencies 

when multiple cluster nodes run on 

virtual machines on the same hardware. 

Let us assume we run a virtualized architecture 

on the blades that do not have AMC cards. In 

this case we have a virtual machine monitor 

and a number of virtual machines managed by 

each monitor. This is reflected in the information 

model as shown in figure 3. This hierarchy 

is then combined with the tree of hardware element 

objects. Additional dependencies, for instance 

dependencies of an object representing 

an operating system to an object representing a 

disk on a different compute blade, can also be 

defined in the PLM information model. Other 

AIS services as well as applications can apply 

PLM managed objects to their interactions and 

dependencies to hardware, virtualization and 

operating system layers. This starts by mapping 

cluster nodes of the SA Forum cluster membership 

service (CLM) to execution environments 

in the same way as the availability management 

framework (AMF) maps its entities to 

the cluster nodes. The SA Forum AIS information 

model now provides a complete representation 

from hardware to application services 

(figure 4). With PLM objects, the information 

model can now easily indicate which AMF 

components are running on which blade, even 

in case of virtualization. At the same time it allows 

high flexibility. There may be execution 

environments which do not run cluster nodes, 

but only software entities that are not aware of 

the SA Forum middleware. Configuration of 

services can be done without knowing whether 

there is a virtualized architecture on this hard- 


ware or not. PLM uses its objects in the information 

model to control operation and administration. 

It provides administrative commands 

for these objects and uses the objects to 

locate errors and inform users about physical actions 

of the operator, such as extraction and insertion 

of field replaceable units. The information 

model directly reflects the physical structure 

of the hardware. Since PLM objects are used to 

controlallactionsanoperatormaytakewhile 

maintaining the system (for instance, extract a 

piece of hardware), all field- replaceable units 

should be modeled as PLM objects. The system 

architect is free to model a finer granularity, i.e. 

create objects for hardware entities that are not 

separately replaceable. This allows the operator 

to separately manage sub-components, if HPI 

supports the operations. On the other hand, if 

the operational rules and procedures do not 

allow exchanging a certain entity, no object is 

needed to represent that entity in the information 

model. PLM will then assume errors of the 

entity as errors of its parent. 

The PLM service will inform its users using 

callbacks if the operator starts physical actions. 

In the same way, it will inform users about administrative 

operations. This is done using a 

track interface as is usual for AIS services. Some 

events allow graceful termination of all affected 

services, but other events do not. In case of 

a lock command or if the latches of an entity 

are opened, services can be gracefully terminated. 

However, this graceful termination may 

not be possible in case of hardware faults. Also 

if the operator does not wait for the blue LEDs, 

and extracts a board before all services are 

terminated, it is important to inform users 

about the so-called surprise extraction. 

Redundancy management in AIS is based on 

state management. State changes are reported 

using notifications, and state changes trigger actions 

on dependent entities. PLM defines state 

models for hardware elements and for execution 

environments. The state model is similar to 

the AMF state model and can be mapped on 

the X.731 state model recommended by CCITT. 

For hardware elements, the presence state reflects 

the HPI hot-swap state for entities that 

support hot swap management. Administrative 

commands control the administrative state, 

which for hardware elements also provides the 

locked-instantiation value which can typically 

be mapped to the power state of the hardware. 

Manipulating the administrative states for 

www.embedded-control-europe.com 

The Online Information Source for Europe`s Embedded Engineers 

• Product News • Technical Articles • Company Profiles • Events

Figure 4. The Service Availability Forum AIS information model now provides a complete representation 

from hardware to application services. 

execution environment objects that represent 

virtual machines allows an operator to remotely 

control virtual machines in a virtualized archi- 

■ BittWare: AMC module with multicore 

DSP from TI 

BittWare announces that they will be supporting 

Texas Instruments’ TMS320C6474 digital 

signal processor on their F1/GX-AMC. Featuring 

the multicore C6474 DSP designed 

specifically for telecommunications infrastructure 

applications and coupled with a 

high-density Altera Stratix II GX FPGA, the 

BittWare F1/GXAM provides a completely 

flexible, reconfigurable platform for high-end 

digital signal processing. 

News ID 897 

■ Schroff: MTCA system for 6 single 

mid-size AMC modules 

For powerful multi-processor applications in 

the smallest spaces, such as are required for 

image and video processing in industry or medicine 

or as media servers in transport applications, 

Schroff has developed a compact MicroTCA 

system. This system contains six slots 

for the horizontal mounting of AdvancedMC 

single mid-size modules and a MCH. The 19" 

wide and 1 U high system, with a depth of 350 

mm, is cooled efficiently by a hot-swapcapable 

fan tray with cooling unit management. 

Air extraction is from right to left and has been 

optimised by the special design of the case. 

News ID 723 

tecture using the PLM service. Hardware faults 

and faults of the operating system layer usually 

cause a number of subsequent state changes 

Product News 

■ N.A.T.: adaptor integrates PCI cards in 

MicroTCA systems 

N.A.T. introduces its newest plug-and-play 

member of the family of adaptors at electronica. 

XLINK allows the easy and direct integration 

of PCI based systems into MicroTCA 

based systems, providing all advantages of a MicroTCA 

system without any additional software 

effort. XLINK comprises of a pair of an AMC 

module and a PCI card, connected by a PCIexpress 

link cable. The AMC module resides in a 

MicroTCA system and the PCI card is plugged 

into the system-host slot of a passive PCI 

backplane. From the perspective of system 

host in the MicroTCA system the PCI sub-system 

as integrated by XLINK, is considered to be 

just another AMC module with multiple PCI 

devices on it. 

News ID 720 

■ RadiSys expands AdvancedTCA 

portfolio 

RadiSys has expanded its AdvancedTCA portfolio 

with the launch of two new products, the 

Promentum SYS-6016 and SYS-6002 platforms. 

The 16-slot, 13U Promentum SYS- 

6016 platform meets the carrier-grade requirements 

of the telecommunications industry. It 

comes standard with the Promentum ATCA- 

2210 10-Gigabit switch and control modules 


in the system. For instance, when an operator 

starts to remove replaceable units from the 

system components are terminated; services 

fail-over or even terminate. In a live system, it 

is important to be able to identify the root causes 

of these events. Typically redundancy is used 

to avoid a hardware fault causing a service outage. 

However, service outages sometimes occur, 

and in that case it is very important to identify 

the root cause of a failure. With the PLM 

service in place, an AIS implementation can 

correlate state changes with root causes like 

hardware failures, physical interaction with 

the hardware (for instance hardware removal) 

or administrative commands. 

As in the previous example, we again call to 

mind the AMC disk modeled as a managed object. 

Let us assume that the disk contains the system 

partition of a UNIX-type operating system 

running on another blade. On top of this operating 

system, we have a cluster node, AIS services 

and applications. The disk may fail or run out 

of free space. In this case, the PLM service will 

often succeed in gracefully terminating all services, 

despite the faulty disk. State changes of the 

services and applications, even a service outage, 

can therefore be correlated to the faulty disk. ■ 

which provide highly integrated centralized 

functions such as switching, shelf management, 

network-timing and system management 

capabilities. 

News ID 709 

■ BPM: support for embedded USB host 

devices from FTDI 

BPM Microsystems announces support for 

Future Technology's Vinculum VNC1L 

embedded USB host controller IC. Aimed at 

engineers designing Vinculum VNC1L applications, 

BPM Microsystems device programmers 

are also suitable for programming service 

providers, contract manufacturers and 

component distributors with option of equipment 

from basic manual programming systems 

to fully automatic bulk pre-programming. 

News ID 942 

■ PT: communication controllers in 

AMC form factor 

Performance Technologies announces the availability 

of two new communication controllers, 

the AMC304 and AMC305, for deployments in 

telecommunications, aerospace and defense, as 

well as commercial markets needing time-division 

multiplexing communication controllers 

in an AMC form factor. 

News ID 768 

33 February 2009

SMALL FORM FACTOR BOARDS 

Pico-ITX single board computers 

combine advantages of COMs and SBCs 

By Robert Kuo, VIA Technologies and Colin McCracken, SFF-SIG 

This article introduces 

the new small form factor 

Pico-ITX for single board 

computers featuring the 

SUMIT interface which 

includes various 

future-proof busses. 

■ Before describing the advantages of Pico-ITX, 

it is necessary to define what is meant by the 

terms SBC and COM. An active backplane SBC 

contains CPU and I/O subsystems as well as I/O 

connectors (PC-style molded connectors and 

pin headers), all in just one computer board. 

The SBC is also known as an embedded motherboard. 

The COM-based approach splits the 

equivalent circuitry and connectors between 

two boards: The COM CPU and the I/O carrier 

board. Usually the carrier board is custom for 

each system manufacturer and application. 

The system OEM is responsible for the COM 

working with the carrier, which is both a curse 

and a blessing since there is greater flexibility 

with I/O expansion and connector location on 

COM carriers. But if the COM doesn’t boot in 

design or on the production line, guess who 

must figure out why? 

An SBC is connected directly to a power supply 

by a cable, while a COM is powered through its 

carrier board interface. In a pure COM architecture, 

the only connectors on the COM are 

the board-to-board connectors between the 

CPU and the carrier; therefore all power, buses, 

and I/O signals must pass through this interface 

to get to the carrier board. Even interfaces that 

are normally cabled, such as disk drives and 

LCDs, are routed through the carrier board 

first. In many high-performance systems, the 

processor and chipset consume most of the 

power budgeted to the electronics, so it seems 

counterintuitive to send so much current 

through tiny high-density surface mount pins 

that attach the COM to the carrier. Thermal solutions 

are usually complete on SBCs (heatsink 

with or without fan). On the other hand, 

COMs offer incomplete thermal solutions such 

as heat spreaders. With or without heat spreader, 

heat must be actively removed or transferred 

to an outside enclosure. Either way, thermal design 

is necessary. COMs offer an illusion of interchangeability 

between module suppliers, 

i.e., second sourcing. However in practice, interfaces 

leave wiggle room for vendor-specific 

implementations, and the stiffness of the 

module power supply or rise time required to 

boot is rarely specified in supplier documentation. 

In most cases, vendor documentation 

doesn’t specify operation in enough detail in 

these areas, so the customer is forced to buy the 

hardware and assess these items empirically. 

Applications that require desktop-style interfaces 

without special I/O might be able to use a 

Pico-ITX product as-is. For mid-range signage / 

advertising or kiosk applications, or for rugged 

portable computers or user interface controllers, 

the requirements include LCD, SATA 

drive, network interface (LAN or wireless), USB 

ports, keyboard / keypad, and sometime MPEG 

acceleration, touch screen or serial ports. Pico- 

ITX fits the bill. A two-board COM solution 


would add unnecessary costs. If an off-the-shelf 

Pico-ITX board doesn’t quite have the right I/O 

or connectors, or has too many features for a 

given high-volume application, simple board 

layout changes would suffice. A single board solution 

would cost less to manufacture and assemble 

without the extra PCBs, connectors, and 

integration. Existing Pico-ITX products can be 

depopulated, i.e. unused components removed. 

If that doesn’t quite get there, minor re-designs 

are still less costly than a COM carrier board 

design because the bulk of the circuitry has 

been proven already in hardware and software. 

Figure 1. Single-row connector of the SUMIT 

interface

Table 1. Development tasks for typical COM-based and SBC-based systems 

For the rest of the low power market, special 

I/O is required, especially in industrial, process 

monitoring, and data acquisition applications 

that use A/D sampling, serial ports, or CAN 

Bus, for example. The Pico-ITXe standard 

contains the SUMIT interface for easy connectivity 

of low-speed I/O. 

At a first approximation, a COM on a carrier 

board can be thought of as an inverted stack of 

an SBC with a custom I/O card (mezzanine I/O 

card) on top. The COM is nearly always smaller 

than the carrier it plugs into, while the SBC 

is large and the I/O card is smaller. Recall, however, 

that the SBC already contains the full 

power supply and filtering circuit whereas the 

COM carrier bears much of the power supply 

burden. In addition, SBCs often come with 

legacy I/O built in, while the latest COMs are 

legacy-free, sacrificing even the most basic element 

of the industrial system – the serial port. 

It turns out that initializing serial ports is a 

BIOS function, so adding serial ports on a 

COM carrier often requires a custom BIOS. The 

BIOS firmware resides on the COM, so this 

reduces vendor interoperability, and if the 

production volume of the application is below 

several thousand per year, the OEM may not get 

the support of custom BIOS at all. Even most 

trade groups have backed away from this challenge. 

Current Pico-ITX boards not only solve 

the power supply ramping issue, but also provide 

legacy interfaces like PATA (IDE), serial 

ports, and PS/2 keyboard and mouse. Adding 

the custom I/O is a simple matter of attaching 

to the flexible SUMIT interface, with PCI 

Express, USB, LPC, I²C, and SPI choices. All of 

this flexibility comes without supplier and 

third party design house finger pointing. For 

SFF SBCs, the problem quickly becomes how to 

bring the I/O off the board without a nightmare 

of cables and spring-board interface cards. 

Pico-ITXe is as small as SBCs come, so the 

challenge is stronger than ever. Previously, the 

approach was multiple rows of pin headers that 

presented system assembly techs with an 

installation nightmare. Subsequently, the pin 

headers were aligned and spring boards 

emerged, but the mechanical support for ends 

of the spring boards was challenging, and 

there was no real savings in the system footprint 

compared to a larger single board. 

The solution involves a clever tiered architecture, 

with just a single row of pin headers on 

only two of the four edges of the board, and the 

rest of the buses and signals consolidated into 

a very high density board-to-board interface 

called SUMIT (pronounced sum-it, which 

implies the addition of I/O). The connectors 

occupy very little board space, leaving room for 

other features. 

Expansion bus connectors are relatively expensive 

in terms of component cost and board 

space occupied, so cost-sensitive applications 

don’t have the luxury of excessive connector 

pins that won’t be used. Most COM form 

factors, like ETX, XTX, and COM Express, were 

architected for 3-chip x86 solutions three to 

seven years ago. The prevalent interfaces back 

then – EIDE (PATA), parallel PCI Bus, parallel 

port, serial ports – are being phased out in favor 

of more space-efficient high-speed serial inter- 


for 

Lowest Power to 

Highest Performance 

Embedded Module Family 

Ultra Low Power 

conga-CA 

with Intel ® Atom Z5xx and 

Intel ® System Controller Hub US15W 

Low Power 

conga-CA945 

with Intel ® Atom N270 and 

Intel ® 945GSE Chipset 

Economy 

conga-B915 

with full feature set & economically 

priced Intel ® Celeron processor 

Performance 

conga-B945 

up to 2.16 GHz Intel ® Core2 Duo 

High End 

conga-BM45 

Intel ® Core2 Duo up to 2.53 GHz, 6 

MByte L2 cache and 8 GByte DDR3 

RAM, high performance integrated 

graphics engine 

03.03.-05.03. 

Hall 12 - Booth 122 

Find more details at: www.congatec.com 

Tel. +49 (991)2700-0 • info@congatec.com 

Intel and Intel Atom are trademarks of Intel Corporation ind the US and other countries. Highscore 

35 February 2009


Figure 2. Location of 60x72mm I/O Mezzanine Module on top of Pico-ITXe (dashed line) 

faces such as SATA, PCI Express, and USB ports. 

SUMIT contains PCI Express lanes, USB ports, 

the Low Pin Count (LPC) Bus, I²C / SMBus, 

and SPI / μWire with several decodes available. 

This interface is a major breakthrough on 

many levels. On the one hand, the cabling 

45nm Core 2 Duo technology 

introduced in embedded COM modules 

By Konrad Löckler, MSC 

■ In the past few months the latest processor 

technology from Intel, forming part of the Intel 

Centrino 2 mobile platform, has enabled new 

laptop models to work faster while consuming 

less power. Additionally, industrial customers 

have received the first samples of COM Express 

modules based on this new technology. As soon 

as Intel announced the latest Core 2 Duo 

processor platform in the miniaturized SFF 

(small form factor) packaging, the first products 

using this chipset/processor combination were 

introduced for the embedded and industrial 

computer market. The trend is for more compact 

and power-efficient solutions to replace the 

older generations of industrial Computer-on- 

Modules. In the lower performance range this is 

nightmare is alleviated. The I/O on the expansion 

module is elevated and recessed from the 

pin headers on the Pico-ITX board, greatly simplifying 

assembly. Much more I/O is possible 

with this architecture by making use of the 

three-dimensional space without growing the 

being achieved by new designs using Intel 

Atom processors based on 45nm technology, for 

the higher end the latest 45nm Core 2 Duo and 

Celeron processors are available. 

MSC has now introduced the MSC CXC- 

GS45. This COM Express module in the compact 

form factor offers, as an industry first, the 

computing performance of high-end notebooks 

on a module only 95x95m² in size. With 

a broad selection of processors, the new module 

family offers performance for many different 

applications. The Intel GS45 graphics and 

memory controller hub (GMCH, 27x25mm²) 

in SFF package is used together with the enhanced 

IO controller hub Intel ICH9M (16x16 


72x100mm base footprint. Secondly, the interface 

is easy for off-the-shelf I/O vendors to design 

to, setting the stage for a growing I/O ecosystem 

for years to come. SUMIT also contains PCI Express 

clocks which will be needed for Gen2 devices 

in the future. Surprisingly, most COM standards 

do not include the clocks since today moderate-speed 

devices can derive the timing from 

the incoming Rx+ and Rx- signals. High speed 

devices (5GHz) will need the separate clock signals 

in the future, and SUMIT is prepared. It is 

also ready for 5 Gigabit USB3, while most 

COMs are unproven for it. Finally, the interfaces 

that comprise this interface are exactly the ones 

that the leading chipset vendors have decided to 

standardize on for their future roadmaps. The 

tight mapping of chipset I/O and buses means 

that no space or cost is wasted, as opposed to the 

large and expensive COM connectors that are required 

to bring out everything from the higher 

power 3-chip x86 solutions. 

With Pico-ITXe, PATA is supported directly 

with the inexpensive and yet rugged 2mm 44pin 

industry standard pin header, without 

having to occupy expensive COM connectors 

and their mating connectors. As PATA is graduallyreplacedbySATAovertime,boardscan 

depopulate or eliminate the PATA connector, 

freeing up board space without forcing a 

change upon a COM interface. ■ 

mm²) . All processors in the latest Core 2 Duo 

45nm generation (former code name Penryn) 

which are available in SFF packaging 

(22x22mm²) can be combined with this chipset. 

The MSC module comes with all processor 

variants that Intel offers with long term availability 

on their embedded roadmap. Specifically 

these are the Core 2 Duo SV (standard 

voltage) SP9300 with 2.26GHz and a TDP 

(thermal design power) of 25W, the Core 2 Duo 

LV (low voltage) SL9400 with 1.86GHz (TDP 

17W) and the Core 2 Duo ULV (ultra low voltage) 

SU9300 with 1.20GHz (TDP 10W). In addition 

the ULV Celeron 722 with 1.20GHz and 

only 5.5W TDP rounds off the selection at the 

lower end.

With the introduction of this 45nm processor 

generation, Intel has achieved major improvements 

in computing performance, graphics acceleration 

and power consumption. The 45nm 

HKMG semiconductor manufacturing technology 

used reduces leakage currents, which 

contributes significantly to reduced power 

consumption. Most importantly the fast front 

side bus with up to 1066GHz, the larger L2 cache 

with 6Mbytes, and the integrated mobile Intel 

graphics media accelerator 4500MHD with up to 

533MHz clock frequency enable up to three times 

the performance of the previous platform. Further 

reduction in power consumption has been 

attained by various improvements in power 

management. In this context dynamic acceleration 

technology should be mentioned, which allowsthesecondcoretobehaltedforpowersaving 

purposes when running single-threaded 

programs. Also the new C6 deep power down 

mode, which switches off cache memory in the 

processor idle state, improves the power equation 

as does the possibility to dynamically switch the 

FSB clock between 533 and 1066 MHz. 

The new graphics controller inside the GS45 allows 

distortion-free playback of high-definition 

videos using hardware-accelerated MPEG decoding 

without any significant increase in 

processor load (Intel clear video technology). 

Two displays with different content are supported 

by the dual independent display mode 

with a maximum resolution of 2048x1536 

pixels. The new graphics core now also provides 

DirectX10 and OpenGL 2.0 support. The rendition 

of high-definition contents is complemented 

by the Intel high-definition audio interface. 

Besides the increase in efficiency to be 

expected with a new Intel platform, the MSC 

CXC-GS45 module includes a variety of bus 

and I/O interfaces. In addition to five PCI Ex- 


Figure 1: The CXC-GS45 is based on the small form factor components of the Intel Centrino 2 

platform. 

press x1 lanes (a sixth is used on the module for 

the Gigabit Ethernet controller), a PCI Express 

x16 (PEG) interface (multiplexed with 2x 

SDVO or video capturing), the PCI and LPC 

bus, eight USB 2.0 ports and a 10/100/1000base-TX 

Ethernet interface are all available. This 

is supplemented by an analogue VGA and two 

24-bit LVDS channels. For data storage devices, 

four SATA2 channels with up to 300MB/s and 

an enhanced IDE port (ATA/UDMA166) are 

present. The DDR2-SODIMM socket can be 

populated with up to 2 Gbytes PC2-6400 

(800MHz) SDRAM DDR2 memory. 

MSC also includes all the quasi-standard features 

already available on previous generation 

modules – such as the connector for a processor 

fan, a programmable watchdog, the backup 

E²PROM for setup data protection, or the 

TPM component from Infineon. The BIOS 

based on the SecureCore line from Phoenix 

Technologies will optionally be available with 

Secure Boot extensions from MSC, which are 

compliant with the Trusted Computing Group 

(TCG) requirements. 

With a typical power consumption of less than 

10W, the Celeron model will be a solution for 

small fan-less systems in industrial or mobile deployments 

requiring medium performance. 

For higher performance requirements in such 

applications, the version based on the SU9300 

presents the best- suited solution, generating 

only a few watts more power dissipation but 

nearly doubling computing performance. If active 

cooling is possible, the system designer can 

also have recourse to the variants with the highest 

computing power (SL9400 and/or SP9300). 

MSC can support the customer with different 

heat spreader or heat sink solutions, which are 

optimally adapted for the modules. ■ 


Unlimited 

freedom 

TQMa31 

The Multimedia Module 

• CPU: MCIMX31 from Freescale 

• FPGA on module 

• Integrated graphic engine 

• USB 2.0 and 1.1 OTG 

• OS: WinCE, Linux 

• 75mmx60mm 

� Extended temperature range 

� Low power consumption 

� Long term availibility 

� Flexible expansions through FPGA 

� Perfect fi t for display-solutions 

» ARM11 TM Architecture 

Visit us: Hall 12, Booth 342 

Email: info@tqc.de 

Phone: +49 8153 9308-333 

www.tq-group.com

RUGGEDIZATION 


technology in industrial computing 

By Abel Lee, Hunter Lin, and Bruce Chen, Moxa 

Wide temperature capability is 

a vital feature for industrial 

computers deployed in 

locations with harsh 

environmental conditions. This 

article explains the design 

technology needed for 

wide temperature industrial 

computers. 

■ Typical industrial standards call for an operating 

temperature range of 0 to 40°C. Although 

these standards are far more rugged 

than those for consumer products, industrial 

computers need higher quality construction to 

provide protection from the elements for use in 

harsh environments. 

Wide temperature industrial computers are 

even more rugged and generally have an operating 

temperature range of -40 to 75°C. The 

ability to operate under such extremes allows 

wide temperature industrial computers to be 

deployed in many more locations, such as 

desert and polar regions. By constructing an industrial 

computer to be operable under extreme 

heat and cold, manufacturers can assure reliability 

and reduce the likelihood of product failure 

for applications in a variety of harsh environments 

over extended periods of time. 

Industrial applications are found in various settings, 

including factories, power stations, roadside 

traffic control boxes, as well as marine, 

desert, and polar locations. Given the severe and 

harsh conditions often found in these environments, 

computers used in these applications 

must be able to endure extreme temperatures. 

For example, a computer server may be deployedinafactorywhereitislikelytobeexposed 

to a great deal of heat. If it is deployed in 

a remote outdoor location, the computer would 

need to be able to operate under extreme cold 

or heat, depending on the climate. A computer 

without wide temperature design is usually 

ill-suited for the harsh demands of industrial 

applications. As a result, wide operating temperature 

has become an important feature for 

industrial computers. 

Onboard fans are the traditional cooling 

method for computers but may burn out and 

cause additional problems. When the fan stops 

operating, the whole system usually crashes. 

The fanless design has become standard among 

industrial computers as it eliminates mechanical 

problems caused by onboard fans. However, 

the fanless design presents additional challenges 

for developers of wide temperature 

products. Not only do the critical components 

selected need to meet wide temperature 

requirements, but the hardware layout design 

that integrates all peripheral computers must 

fulfil wide temperature demands as well. High 

temperatures can have crippling effects for the 

entire computer system. The absence of a welldesigned 

ventilation system may cause the system 

to crash when exposed to prolonged periods 

of extreme heat, increasing system recovery 

effort and maintenance costs. Unfortunately, 

the components that support the optimal system 

performance are likely to generate more 

heat. For example, a high performance CPU 

produces more system frequency and heat. This 


Figure 1. The H-type heat sink 

by Moxa directly contacts the 

main sources of heat. 

challenge is a nightmare for hardware designers, 

since it takes a great deal of effort and time 

to choose the optimal components and lay out 

the hardware to ensure that the computer can 

endure high temperature conditions. In addition, 

we also need a better overall thermal solution, 

which involves placing the components 

on the PC board in an efficient manner, and 

choosing materials with high thermal conductivity 

and an optimal heat sink design. By making 

intelligent use of standard thermal conductivity 

theory, we are able to lower the thermal 

resistance ( ) at each point on the heat 

transfer route. 

Building an industrial computer that can operate 

under low temperatures poses another 

challenge to manufacturers as they usually 

focus more on addressing the high temperature 

issue. Computers always generate heat when 

they run, so it is understandable for manufacturers 

to be more concerned with developing an 

optimal cooling system. However, some industrial 

applications take place in cold climates and 

computers used in these settings need to be 

heated in some way. Generally speaking, the 

better a computer is at dissipating heat in high 

temperature environments, the less effective it 

is in low temperature applications. The more effort 

manufacturers invest in addressing high 

temperature issues, the more effort they need to 

invest in addressing low temperature issues as

well. Costs are among the biggest factors manufacturers 

consider when designing a new 

product. Wide temperature design requires 

higher quality components and more effort in 

hardware layout and system integration. 

As a result, it is more costly to develop and produce 

wide temperature computers, which 

makes the products less competitive in terms of 

price. Lowering production costs while enhancing 

the value for wide temperature products 

has become a critical concern for industrial 

computer manufacturers. Designing a wide 

temperature industrial computer requires a full 

understanding of the product thermal gradient 

in order to optimize the placement of components. 

Several factors need to be considered 

with regard to the thermal placement of com- 

RUGGEDIZATION 

Figure 2. The simple cooling method of vent holes does not prevent dust or water from entering 

the computer. 

Figure 3. Use of a heat pipe to conduct heat out of the computer. 

ponents inside the computer. First of all, hardware 

engineers need to identify the main heat 

sources and hot spots so layout designers can 

optimize the component placement on the 

motherboard. Basically, the closer a component 

is to the main heat source, the more durable it 

needs to be. Designers can also reduce the number 

of heat sources by using components that 

generate less heat and arranging the components 

in the most optimal positions. 

The chassis and total system power consumption 

should also be considered when developing 

a wide temperature computer. For example, 

using a larger chassis or reducing the system 

power consumption can help dissipate the 

heat generated by the computer. It is also crucial 

for engineers to determine a main direction 

39 February 2009

RUGGEDIZATION 

Figure 4. Thermal cycle inside a heat pipe. A. Fluid turns into vapor absorbing thermal energy. B. 

Vapor migrates along the cavity to the end at a lower temperature. C. Vapor condenses back to 

fluid and is absorbed by the wick, releasing thermal energy. D. Fluid flows back to end at a higher 

temperature 

for heat transmission. This involves a sophisticated 

understanding of the component placement 

and a technical arrangement of the components 

to disperse the heat via a specific transfer 

route. Understanding the system thermal 

gradient is essential to optimizing the thermal 

placement of components and designing wide 

temperature computers. 

Environmental test chambers are an important 

way to determine if a product can be used in 

harsh surroundings. Most manufacturers use 

forced-convection thermal chambers for testing. 

However, results from these tests are usually 

unreliable as the environments they replicate 

are generally inconsistent with actual environmental 

conditions found in industrial applications. 

Using a natural-convection thermal 

chamber allows engineers to establish a windless 

environment that more closely resembles 

actual industrial application settings. 

Using wide temperature components is the 

most direct way to produce wide temperature 

computers. To make it easier to find and deploy 

wide temperature components, hardware and 

layout designers should construct a database of 

components that meet the rugged requirements 

for use in wide temperature environments. 

Testing components, materials, and products in 

a natural-convection thermal chamber first 

Figure 5. Heat sinks are another solution for 

transferring heat. 

makes it easier to determine which ones are 

suitable for the wide temperature database. This 

database is extremely important and helpful 

should you decide to convert a standard temperature 

product into a wide temperature one. 

Designers can easily choose the components 

from the database and deploy them in the product, 

which accelerates product development 

and shortens time-to-market. 

The simplest way to transfer heat from the 

computer is to use vent holes. These holes are 

usually arranged in pairs and located on both 

sides of the computer (often vertically) so that 

hot air can flow through the computer and accelerate 

the ventilation effect. However, this 

method is only suitable if the computer does 

not generate too much heat and there is good 

airflow at the field site. Unfortunately, this 

method cannot prevent dust or water from 

entering the computer. 

Another method is to use a heat pipe to direct 

the heat out of the computer. This solution employs 

specific materials and instruments to 

transfer heat via the thermal cycle. It is particularly 

suitable for board-based products that 

contain CPUs and chipsets as the primary heat 

sources. For example, silicon thermal pads can 

be used to directly cover the onboard CPU and 

chipsets, which are then covered by aluminium 

heat absorbers located on the slat where one or 

more bronze heat pipes are affixed. The heat 

pipes are hollow but lined with a wick containing 

a working fluid, such as water, that can 

absorb the heat from the heat sources. The 

pipes are often led to a location where it is easy 

to dissipate the heat, such as a plate at the front 

of the computer. In addition, the main function 

of the heat pipes is to transfer heat from one 

side of the computer to another via the thermal 

cycle. 

Heat sinks are another solution that can be used 

for transferring heat. The heat sink is usually 

made of bronze or aluminium, materials that 

can easily conduct heat from one side to an- 


other. For example, a heat sink made of aluminium 

may be used to directly cover the component 

and absorb the heat to transfer it out of 

the computer. This method is widely used and 

can serve as the cooler for the CPU or the entire 

computer. It usually has a fin-shaped design 

to maximize the surface area and speed up the 

heat transfer. When a heat sink is used, the size 

makes a big difference since the designer needs 

to optimize the sink to maximize the heat transmission 

effect. However, traditional heat sinks 

can only solve a limited number of challenges 

as they are externally attached to the unit. It is 

not guaranteed that all the heat can be directed 

to the sink and lower the temperature of the 

computer. To solve this problem, Moxa has introduced 

a patented H-type heat sink design to 

reduce the internal temperature. H-type heat 

sinks involve a plate that can be inserted into 

the inner part of the unit to make direct contact 

with the main heat sources. This design ensures 

that most of the heat can be absorbed from the 

main heat sources and directly transmitted to 

the external fin-shaped plates. This solution is 

also more cost-effective compared to the 

traditional heat sink design. 

The methods mentioned are all geared towards 

heat transmission and only apply to high temperature 

environments. However, different 

technology is required to ensure reliable operation 

for computers used in cold climates and 

settings. Balance heaters that automatically 

start working when the exterior temperature 

drops too low can be used to warm the interior 

of the computer. However, this method 

requires precise and accurate temperature 

configuration to ensure that the resistors start 

operating when needed. 

Wide temperature industrial computers generate 

less heat and consume less power, making 

them particularly reliable and less prone to failure 

over long periods of time. As a result, they 

are well-suited for the following situations. Outdoor 

applications, such as deserts or mountains, 

where it is difficult or costly to build a climatecontrolled 

shelter for sensitive electronic equipment. 

Indoor applications, such as a factory 

floor, where equipment must be placed near 

machines that generate extreme heat. 

Mobile or mixed applications in harsh environments, 

such as in the military, where machines 

must operate reliably in low and high 

temperatures. Wide temperature models present 

a much more reliable and affordable alternative 

to using regular industrial-grade devices. 

They are an ideal solution for any application 

involving harsh industrial environments, such 

as power substation automation, gas stations, 

intelligent transportation systems, environmental 

monitoring, factory automation, and 

other related systems. ■

RUGGEDIZATION 

Customized CompactPCI CPU boards 

for industrial image processing 

By Ingrid Einsiedler, Kontron 

Sophisticated image 

processing systems are 

becoming indispensable for 

quality control on automated 

filling and packaging lines 

with zero failure tolerance. 

A leading manufacturer of 

filling lines and control 

systems has developed a 

highly flexible image 

processing system based on 

customized 3U CompactPCI 

CPU boards. 

■ Based in Neutraubling, Germany, Krones is 

the world’s market leader in supplying systems 

and complete processing lines for filling and 

packaging. The company designs, develops, 

produces and installs customized filling lines for 

beverage manufacturers and the food packaging 

industry, as well as the chemical, pharmaceutical 

and cosmetic industry. An increasingly 

indispensable part of modern processing 

lines is the image processing technology that is 

used for quality control at various stages of the 

production and filling process, e.g. after washing 

to check the quality of the outside and inside 

of containers, after filling to check the filling 

level and the seal, and after labelling to 

check the position of each label and the “best 

before” date. Depending on the number of 

stages required, a typical filling line could have 

as many as 20 different camera systems each 

running different software applications. 

With a large number of camera systems running 

various applications in different industries, 

the challenge for Krones was to develop a costeffective 

system that is easy to integrate, install 

and maintain regardless of the application 

area. Reliability is also critical – the systems 

needtooperate24hoursaday,7daysaweek 

under tough operating conditions with strong 

vibrations emanating from the conveyors and 

other plant machinery. To meet these demands, 

Krones developed a universally appli- 

cable system with a central server and distributed 

image processing units based on customdesigned 

CP306 CompactPCI CPU boards 

from Kontron. The CompactPCI boards provide 

exactly the right cost-effective performance 

and high availability combined with longterm 

availability. 

CompactPCI is a natural choice for the harsh 

operating conditions found on the high-speed 

processing lines in various industries. The robust 

connectors, metal guides and metal front 

plate hold the CPU boards firmly in place inside 

the camera system mounting case. The fanless 

design with a heatsink directly screwed to the 

board, as well as directly soldered CPU and 

memory, provide a board design that is inherently 

resistant against vibrations and shocks. 

Moreover, the CompactPCI connector offers 

optimal protection against the damp conditions 

and, in some cases, the aggressive airborne 

agents found in filling plants. 

To keep production and maintenance costs to 

a minimum, Krones devised a novel blank slate 

concept that enables image processing units 

with the same hardware configuration to be deployed 

on a wide range of different applications. 

The Krones inspection and control system 

consists of a central server connected via fast 

Ethernet to a number of distributed image processing 

systems located at various stages of the 

Figure 1. The image processing 

system based on custom CP306 

CompactPCI CPU boards is 

suited for zero-failure tolerance 

filling lines with rugged, harsh 

environments. 

production line. Each image processing system 

consists of up to four digital cameras (depending 

on the application), the corresponding 

number of frame grabber cards and a customized 

Kontron CP306 CompactPCI CPU 

board. After pre-processing the information 

from the cameras, the frame grabbers transmit 

the uncompressed image data, which could be 

a simple black and white bit-map, a grey-scale 

image or an RGB image depending upon the 

application, to the Kontron CP306 via the 

board CPCI connectors. Image processing software 

written by Krones runs on the CP306 

under Linux OS and determines, for example, 

whether a container is clean or contaminated or 

whether the fill level is within specified limits, 

depending on the designated task. The result 

(good or defect) is transmitted back to the 

frame grabber and forwarded via Ethernet to 

the central server, which is responsible for highlevel 

control of the processing line. 

The CP306 boards boot via Ethernet from the 

central server from which they also receive the 

application software and parameters for the required 

image processing task. This system concept 

makes the Krones inspection and control 

system extremely flexible and cost-effective. The 

individual image processing systems are like 

blank slates, ready to perform whatever image 

processing task is required. They are therefore 

application- independent and can be used for 

41 February 2009

RUGGEDIZATION 

Figure 2. A high-end blank slate system with a 

total of 8 camera systems on two levels 

Figure 3. Image processing system with four 

CP306 boards connected to multiple framegrabber 

cards 

a wide range of different inspection tasks at various 

stages of the production line. This brings 

a number of important time and cost benefits. 

An image processing system used for reading labels, 

for example, can be quickly and cost-effectively 

dismounted and integrated into an 

earlier processing stage where it might perform 

a completely different image processing task. 

Moreover, the blank slate concept simplifies 

configuration of the image processing systems 

since there is no need to configure the systems 

for individual applications. Krones simply assembles 

the systems with the appropriate number 

of cameras and frame grabbers and a 

CP306 CPU board. 

This enables to efficiently manufacture batches 

of systems every year with minimal integration 

costs regardless of whether they are used 

for measurement, surface inspection, texture inspection, 

code reading, text reading, print and 

inscription control or behaviour analysis. It also 

ensures that replacement systems are always on 

hand in the event of a fault because there is no 

need to supply an application-specific replacement. 

Another critical cost-saving factor was 

the ability to supply a custom version of the 

CP306 CPU board. For the Krones solution, the 

Kontron team of hardware engineers annually 

supplies a large batch of custom boards reduced 

to only the application-necessary components. 

Since the CP306 boots from the central server, 

there is no need for a battery, clock buffer, 

Compact Flash or hard drive. Unneeded interfaces 

have also been omitted – each system includes 

a specially adapted CP306 CPCI board 

containing only one Ethernet port for communication 

with the central server and 4x 

CompactPCI slots for the frame-grabber cards. 

The Kontron software engineers have also developed 

and supplied a customized BIOS for 

Etherboot 3.6 compatibility. The result is a 

slimmed-down and cost-optimized custom 

design for universal application in the Krones 

system configuration. 

Stable real-time operation is also crucial. Image 

processing needs to keep pace with the speed of 

production. Any containers that are not completely 

processed by all the graphic inspection 

systems are treated as defective and removed 

from the production line, resulting in higher 

costs. The CP306 receives uncompressed image 

frames from the frame-grabber cards. The picture 

repetition rate is usually between 20 and 

40Hz but can reach 100Hz in certain applications. 

Running at 100Hz with 1 digital picture 

per container, the CP306 boards are capable of 

handling an throughput of up to 360,000 containers 

per hour. This level is never reached in 

actual processing lines. All the applications running 

at 100 Hz require more than one digital 

picture per container and the average throughput 

lies between 60,000 and 120,000 containers. 

The CP306 easily handles the average data 

transfer rate on the PCI bus of 60 Mbytes/s. 

Peak load is 80 Mbytes/s. The necessary processing 

performance for the CP306 boards is 

provided by either the 1.5GHz Intel Celeron M 

processor or 1.8GHz Intel Pentium M processor 

depending on the intended applications. 

Using two different performance versions optimizes 

the price-performance ratio of the systems 

while maintaining the high degree of standardization 

required for easy integration and 

deployment flexibility. The result is an extremely 

flexible, robust, low maintenance and 

high-performance system that is running 

smoothly in numerous filling lines all over the 

world. 

In addition to the Kontron CP306 CompactP- 

CI board that offers rugged and cost-effective 

Intel Pentium M processor performance, applications 

requiring greater processing power 

will benefit from the Kontron CP307 and 

CP307-64 that bring dual- core performance to 

3U CompactPCI CPU boards. The CP307-64 is 

the most powerful version with Intel Core 2 

Duo processors. Integrated with the Intel Mo- 


bile 945GM Express chipset and ICH7-R 

Southbridge, the CP307-64 achieves high performance-per-watt 

values in a 3U form factor. 

It has an extensive set of on-board features, including: 

options for 1.5GHz low-voltage 

(L7400) or 2.16 GHz (T7400) processors; 

667MHz front side bus; a DDR2-SDRAM 

main storage with a 667MHz storage sequence 

that is expandable up to 4 Gbytes; 10.6 Gbits 

data throughput; and extensive communications 

interfaces. 

Moreover,thereare2xGbitEthernetconnectors, 

up to 6 x USB 2.0, a maximum of 4 x 

SATA-300 interfaces, and an available CompactFlash 

socket. To complete the CPU architecture, 

PCI-Express is used as the data bus to 

the Ethernet channels for high network bandwidths. 

The graphics accelerator integrated 

into the Mobile Intel 945GM Express ensures 

excellent 2-D, 3-D and video features for VGA 

and DVI, which are available in dual operation. 

The CP307-64 is offered in single slot (4HP) 

and dual slot (8HP) options, with the latter providing 

additional legacy support, namely LPC, 

COM, DVI and 2.5” SATA. Rear I/O variants of 

the board with suitable rear modules allow 

service-friendly system structures in which 

the wiring and interface connections are shifted 

to the rear panel of the system. 

The CP307-64 is suited for a broad range of 

compute-intensive embedded applications. 

With a soldered processor and up to 2 Gbytes 

soldered storage, it can be used under the 

harshest environmental conditions, such as assembly 

lines, aircraft, rail and maritime applications, 

as well as mobile test benches and 

measuring devices. With such high CPU performance 

in a small form factor, the board is 

equally suitable for the high-performance requirements 

of image processing and data collection 

For fastest time-to-market and reduced cost of 

ownership, more and more OEMs are turning 

to vendors who can supply not only a wide 

range of standard boards but, more importantly, 

additional custom design and manufacturing 

services. Kontron already offers the latest processor 

technology on a range of established and 

proven embedded platforms on various form 

factors, and has the widest form factor portfolio 

in the embedded computing industry. Moreover, 

the company is one of the leading vendors 

with the stability to ensure the long-term availability 

of all form factors. In addition to providing 

high-quality embedded computing technology, 

Kontron also offers the full range of 

value-adding design services, from custom 

design boards and embedded platforms to 

complete systems in customized housings, and 

has its own manufacturing facilities in Malaysia 

for high volume production. ■

Mainboard for long-term operation 

under extreme conditions in robots 

To operate the PC-based 

control system of their robots, 

Kuka Robotics needed an 

extremely stable platform. 

The long-term availability of 

components, guaranteeing a 

product life-cycle of up to 

seven years, was vital. Based 

on a high-performance 

workstation motherboard, a 

customised and future-proof 

solution was developed in 

cooperation with Fujitsu 

Siemens Computers. 

■ With approximately 2,000 employees working 

in 25 branch offices worldwide and an 

annual turnover of 400 million euros, Kuka 

Robotics is the leading supplier of industrial 

robots in Europe. Their state-of-the-art 

machines are mainly used in the automotive 

and metal processing industries. Among their 

customers are BMW, Daimler AG, Ford, and the 

Volkswagen Group. In 1974, the Augsburgbased 

enterprise presented the first electrically 

operated and microprocessor-controlled industrial 

robot. For more than two decades, the 

solution proved successful on the market, 

earning Kuka international recognition in the 

field of industrial automation. 

A quantum leap in the development of industrial 

robots occurred in 1996, when Kuka Robotics 

presented the first robot control system 

fully based on Windows. By using a technology 

familiar to millions of users from their daily 

work processes, the KR-C1-Control combined 

simple management and flexibility with highest 

quality, performance and complexity. Kuka 

Robotics was the first robots manufacturer to 

replace manufacturer-specific control with a 

standardised and PC-based concept. Respectively 

qualified standard components and 

Kuka-designed housings provided the platform, 

the mainboards used were mainly provided by 

Asian suppliers. “The evaluation of boards 

and providers turned out to be time-consuming 

and cost-intensive”, Josef Leibinger, Manager 

Material Group Management Electronics 

Kuka, remembers. Therefore, the company 

started looking for alternatives in 2004. “We 

searched for a high-performance platform that 

would be available for an unusually long period 

of five to seven years without the need for reorganisation, 

requalification or modification”, 

Leibinger explains. The mainboard - central 

component of the control unit - had to meet 

special requirements, too. In addition to guaranteed 

long-term-availability of all 

components, a reliable system performance 

even under extremely rough and demanding 

conditions was considered essential. 

“Our control systems have to operate 24 hours 

aday,7daysaweekattemperaturesupto 

45°C”, Josef Leibinger adds.“So what we looked 

for were standardised, yet advanced high-end 

RUGGEDIZATION 

Figure 1. Kuka robot used in 

a depalletising application 

motherboards for an industrial product exposed 

to extreme situations and severe conditions. 

Downtimes are in no way acceptable for 

our clients.” During a trial period of several 

months, the solutions of four different 

providers were tested regarding the component 

quality and their compatibility with already existing 

control variants. “Benchmark programs 

provide us with data enabling an evaluation of 

how the mainboard processes our software”, 

Leibinger explains. Special emphasis was placed 

on process response times (track planning, 

graphical display, network communication, 

serial interface communication) and access 

speed to memory components or PCI cards 

(sensors, fieldbusses, inputs/outputs). “Additionally 

to the performance, the focus lay on 

continuous high quality, guaranteed supply 

availability and prices”, Leibinger describes the 

final selection criteria. 

After an eight-month test stage, Kuka selected 

the D1688-K motherboard by Fujitsu Siemens 

Computers. “For us, the entire system is important 

from the quality of the motherboard to 

the chipset up to the circuit board. To ensure 

www.embedded-control-europe.com 

The Online Information Source for Europe`s Embedded Engineers 

• Product News • Technical Articles • Company Profiles • Events 

43 February 2009

RUGGEDIZATION 

Figure 2. Kuka control unit KR C2 for robots 

Figure 3. D1688-K mainboard integrated in 

the control unit 

the design of a standardised product, we need 

a high-quality motherboard produced in large 

quantities that is specially tailored to our 

requirements”, Josef Leibinger says. For Peter 

Hoser, Director OEM Sales Fujitsu Siemens 

Computers GmbH, the development of an embedded 

solution for Kuka was an extremely demanding 

challenge. “Due to the special requirements 

in the processing industry and 

various operating conditions for Kuka clients, 

we needed a high-quality motherboard that is 

completely compatible with a wide range of 

previous systems and able to endure continuous 

operation without downtimes.” 

What was particularly interesting for both 

parties is the different approach regarding the 

usual requirements for mainboards. Leibinger 

explains: “The usual motto of the PC industry, 

“higher, faster, and more colourful”, was inapplicable 

this time, since we had different priorities. 

We needed reliable components with the 

longest possible lifecycle, whereas our demand 

for clock rates, memory upgrade or similar features 

was rather modest.” The challenge for Fujitsu 

Siemens Computers lay in developing a 

special configuration with focus on reliability 

and longevity with moderate requirements for 

technical performance. “In order to ensure a 

familiar and user-friendly environment for 

programming and reliable interaction with 

the robot itself, Kuka required the support of 

embedded Windows XP and the VxWorks operating 

system”, Hoser says. “Since the process 

of integrating and validating version upgrades 

can be costly and time-consuming for industrial 

clients, stable equipment versions are essential 

for Kuka. Additionally, the network 

chip had to be integrated on the motherboard 

delivering a bandwidth of 1 Gbit/s. Real-time 

robot control requires better signal quality 

and latency exceeding by far the specifications 

for standard desktop systems.” To ensure the 

compatibility of selected CPUs and chipsets 

with client systems is an immense investment in 

the fields of research and development for 

Kuka. Any changes during the product lifecycle 

would put the reputation and business of the 

company at stake. 

The component platform equipped with Intel 

technology is part of a long-term roadmap for 

embedded products guaranteeing stability and 

long-term availability.“Kuka doesn’t prioritise 

latest technology, but reliability throughout the 

entire lifecycle and long-term availability of 

essential components,” Peter Hoser explains, 

“and Fujitsu Siemens Computers makes every 

endeavour to guarantee just that. Quality made 

in Germany, the use of high-quality components 

with long-term availability, a consistent 

roadmap and revision process and, last but not 

least, the reliability of an established German 

enterprise are the essential factors that make 

our cooperation successful.” 

Besides the extensive development experience of 

Fujitsu Siemens and its comprehensive system 

competence, the short distance between the 

companies was considered yet another important 

advantage. Leibinger says: “Fujitsu Siemens 

Computer research and development departments 

as well as the production facility itself are 

placed right here in Augsburg - just like ours. 

This locally based technical competence enables 

short-term technical modifications of the mainboards 

at any point in time.” 

The latest control unit generation from Kuka 

has been using Fujitsu Siemens Computer 

mainboards since January 2005, and the summary 

is entirely positive. “Any modifications 

necessary were implemented within the shortest 

time possible and we were able to make use 

of the complete potential for improvements - a 

fact especially important in the start-up phase 

of the project,” remembers Josef Leibinger, 

who has been very satisfied both with the product 

reliability and functionality during the 

complete usage period. Every two years, Kuka 

revises their control technology processes and 

devices – a new challenge Fujitsu Siemens 

Computers is looking forward to. ■ 


Editors 

Jürgen Hübner 

phone +49(0)8092-2477413 

fax +49(0)8092-2477429 

jh@iccmedia.com 

Wolfgang Patelay 

wp@iccmedia.com 

Tony Devereux 

devrex@teyboyz.freeserve.co.uk 

For Reader Inquiries and Address Changes 

please contact: 

info@iccmedia.com 

Sales & Marketing Director 

Manfred Blumoser 

phone +49(0)8092-2477411 

fax +49(0)8092-2477429 

mb@iccmedia.com 

Claudia Mellein 

cm@iccmedia.com 

Christiane Lockner 

cl@iccmedia.com 

Sales Office - UK and USA, Benelux, 

Scandinavia 

Blue Sky Communications 

Malcolm Cameron 

Kent Innovation Centre 

Thanet Reach Business Park 

Millennium Way 

Broadstairs Kent CT10 2QQ UK 

phone +44 (0)1843 609357 

fax +44 (0)1843 609358 

mc@blue-sky-communications.com 

Sales Office - Asia 

Jean Cheng, jean@i-media.com.tw 

Vivian Hung, vivian@i-media.com.tw 

Innovative Media Information & Services 

7F-3, No. 26, Sec. 2, Ming-Quan East Rd. 

Taipei 104 Taiwan 

phone +886 2 2563 1186 

Head Office 

ICC Media GmbH 

Rauwagnerstr. 5 

85560 Ebersberg / Germany 

Editorial Office UK 

36a Blackacre Road 

Theydon Bois 

Essex, CM16 7LU 

Copyright© All rights reserved. 

No part of this publication may be reproduced or transmitted in 

any form or by any means without the prior express written 

permission of ICC Media. 

Although we make every effort to present up-to-date, accurate 

information, boards&solutions will not be responsible for any 

errors or omissions or for any results obtained from the use of 

such information. The magazine will not be liable for any loss 

caused by the reliance on information obtained on this site. 

Furthermore, boards&solutions does not warrant the accuracy 

or completeness of the information, text, graphics in this magazine. 

The opinions expressed in the articles are those of the 

authors and not necessarily the opinions of the publisher.

■ N.A.T. releases new version of 

NATView tool 

N.A.T. has introduced its newest version of 

NATView, the configuration and management 

tool for MicroTCA systems. The current version 

contains further improvements and features 

such as an improved FRU editor, system event log 

including filters and ‘online’ event visualisation. 

NATView is an easy to handle tool for visualising 

and managing any MicroTCA system that contains 

a MicroTCA Carrier Hub of N.A.T. 

News ID 823 

■ Kontron: industrial thin client with 

touch functionality 

Kontron announces its second generation of the 

Kontron Micro Client. By focussing on intended 

use as an industrial thin client including 

touch functionality for operating and monitoring, 

the new Micro Client II provides all the 

latest features required for thin client oriented 

HMI applications in a cost-optimized design 

that includes a modular IP65 protected front 

panel with touch screen. 

News ID 839 

■ Avalue: mini-size dual-core industrial PC 

Avalue’s mini PC EPC-945 is suitable for variety 

of spaces installation with its compact appearance. 

Embedded with an Avalue 3.5 inch 

single board computer, EPC-945 adopts Intel 

dual-core processors and 945GME chipset 

technology, equipped with capabilities of both 

low power and high reliability. Its excellent expandability 

satisfies various environmental requirements 

asked from customers. 

News ID 946 

■ AdaCore announces GNAT Pro for 

Nucleus OS 

AdaCore announces the availability of GNAT 

Pro for Nucleus OS, the embedded operating 

system from Mentor Graphics. Nucleus OS offers 

a highly configurable kernel and utility extensions, 

making it a popular choice for small, highvolume 

embedded computing applications 

where low cost and high reliability are critical. 

News ID 694 

■ Portwell: Atom-based nano-ITX board 

Portwell announces the launch of PVS-1A10, a 

low power, compact DSS system based on 

Portwell NANO-8044 nano-ITX form factor 

embedded board. NANO-8044 embedded 

board is built on a highly integrated two-chip 

solution, the Intel Atom processor Z5xx series 

and Intel System Controller Hub US15W, with 

total TDP less than 5 watts. Combining GADI- 

WA-P0901 industrial supply, fanless NANO- 

8044 embedded board and a standard PCI-Ex1 

card, PVS-1A10 is low power DSS system, 

ideal for small and fanless DSS edge devices 

such as IP Video Server. 

News ID 702 

PRODUCT NEWS 

■ SYSGO: PikeOS is DO-178B reference 

platform for A350 XWB aircraft 

SYSGO announces that their PikeOS safe and 

secure virtualization platform has been selected 

by Airbus for the FSA-NG, a DO-178B certified 

equipment to be deployed on the A350 

XWB aircraft. The purpose of the FSA-NG system 

is to provide the pilot with access to applications 

and information during the flight as 

well as to provide data for the maintenance 

team. The key features and requirements of this 

program include a soft real-time operating system 

capable of supporting ARINC-653 partitioning, 

a reliable, safe and secure dedicated 

high capacity Mass Storage Unit Cabinet, software 

segregation of application services, and 

multi-purpose graphical modules dedicated to 

managing the user's access to applications 

through cockpit terminals. 

News ID 796 

■ Artila: rugged ARM9-based SBC with 

data acquisition 

Artila announces their iPAC-5070, an ARM9based 

single board computer with advanced 

data acquisition capability. The iPAC-5070 integrates 

low power ARM9 SoC, isolated digital 

and analog I/O, 64MB SDRAM, and 16MB 

flash memory with a pre-installed Linux OS 

into a compact and DIN RAIL mountable 

module. Equipped with four channels of 16-bit 

sigma-delta A/D, the iPAC-5070's fully isolated 

design allows accurate measurements within 

harsh and noisy environments. 

News ID 818 

■ NEXCOM: 3.5" ECX/SBC with modular I/O 

design 

NEXCOM has extended its range of small form 

factor embedded boards with the addition of the 

new Intel ECX Form Factor single board computer 

EBC 301. ECX is one of the industries 

smallest and most widely adopted SBC form factors. 

The new EBC 301 features Intel 910GMLE 

and ICH6M chipsets, DDR2 on board mounted 

memory, integrated graphics with DVMT memory 

to support CRT and LVDS function with 

dual independent display capability. 

News ID 786 

■ PT delivers integrated hardware and 

software to CMSI 

Performance Technologies announces that 

Carrier Management Systems has selected 

the company’s PCE385 T1/E1 adapter card 

and integrated NexusWare WAN HDLC software 

to run CMSI’s newly launched NAMS 

Firefly product. NAMS is an integrated hardware 

and software system that gives telephone 

companies total network visibility by 

automating tasks such as data collection, optimized 

network usage, fraud prevention, and 

revenue assurance. 

News ID 685 

45 February 2009

PRODUCT NEWS 

■ congatec: Qseven form factor module 

with Atom Z5xx 

congatec presents conga-QA, the first computer 

module based on the new Qseven form 

factor. The conga-QA is fitted with the latest 

Intel Atom Z5xx range of processors and the 

Intel US15W system controller hub. With a 

typical power consumption of < 5 Watts, mechanical 

dimensions which are scarcely bigger 

than a credit card, integrated battery management 

and ACPI 3.0 power management 

functions, the conga-QA is ideal for all mobile 

applications. 

News ID 736 

■ Concurrent: VME/VXS board utilizes 

2.26 GHz Core 2 Duo 

Concurrent announces their first VME/VXS 

critical embedded product based on Intel’s 

45nm process technology and the latest DDR3 

memory devices. The VX 511/06x a single slot 

6U dual PMC/XMC VME/VXS processor 

board featuring the latest low power high performance 

dual core processor from the Intel 

embedded roadmap - the Intel Core2 Duo 

processor operating at 2.26 GHz or 1.86 GHz. 

The board supports up to 6 Gbytes DDR3-1066 

SDRAM. The VX 511/06x optionally provides 

a VXS P0 connector supporting, on the backplane, 

VITA 41.4 (dual x4 PCI Express) and 

VITA 41.6 (dual 1000Mbps baseband 

IEEE802.3) to provide fast data/control transfer 

between other boards in the system. Commercial 

and extended temperature versions are 

available now and ruggedized, conductioncooled 

or air-cooled (to VITA 47), versions will 

be available shortly. 

News ID 878 

■ Curtiss-Wright: rugged 6U VPX SBC 

with Core2 Duo T9400 

Curtiss-Wright announces the VPX6-1952, its 

first rugged 6U VPX single board computer 

based on the Intel Core2 Duo processor T9400. 

The VPX6-1952 was designed for use as the 

General Processor Modules for use in the Integrated 

Computer System of the U.S. Army's Future 

Combat Systems program. 

News ID 707 

■ DDC: multi-I/O MIL-STD-1553 / ARINC 

429 PCI card 

Data Device Corporation introduces a new 

multi-I/O PCI Card. The BU-65590I card provides 

up to four dual redundant MIL-STD- 

1553 channels, sixteen ARINC 429 receive 

channels, four ARINC 429 transmit channels, 

six user programmable Digital Discrete I/Os, 

two RS-232 Serial I/O Channels, two RS- 

422/485 Serial I/O Channels, and an IRIG-B 

time synchronization input. 

News ID 731 

■ MEN: PICMG 2.30 draft standard 

approved 

The CompactPCI Plus PICMG work group has 

approved PICMG 2.30 PlusIO. This is the first 

of the two planned draft standards for CompactPCI 

Plus. PICMG 2.30 is based on the 

PICMG 2.0 CompactPCI Core specification 

and defines the migration path from parallel 

CompactPCI to the serial CompactPCI Plus. 

For this purpose a new peripheral slot was defined, 

together with a connector for differential 

signals with a fast data rate of at least 10 Gbps, 

to support also the next generation serial buses 

like SATA 3.0, PCIe 2.0, 10GBase-T Ethernet, 

USB 3.0 etc. 

News ID 816 

■ Schroff : 19" frame-type plug-in 

units offer improved airflow 

For its off-the-shelf 19" frame-type plug-in unit 

range Schroff has now developed a new, costoptimised 

family of plug-in units that offer improved 

airflow (up to 65 %), greater robustness, 

better EMC shielding and considerably faster 

assembly, resulting in reduced costs. The aluminium 

profile side wall of the new frame-type 

plug-in units PRO range is significantly thinner 

than that of its predecessor. This both saves material 

and also allows more space for the vertical 

flow of air. In addition, the narrower profile 

makes an additional 1 HP available for regular 

use of the second slot. 

News ID 774 

■ Manhattan Skyline: ARM9-based 

single board computer 

PicoMOD1 from Manhattan Skyline is the first 

product in a new family of single board computers. 

This module (80x50 mm) is based on an 

ARM9 CPU from Samsung and runs at a 

clock speed of 400 MHz. PicoMOD1 offers 32 

MB of Flash memory, 32 MB of RAM memory, 

interfaces for 3 serial lines, 2 USB (host/device), 

I2C, I/O, sd-card and ethernet. Additionally 

it offers the same flexible display interface 

as all products from F&S. All STN-, 

CSTN- and TFT- displays with resolutions up 

to 640x480 can be connected. 

News ID 5 

FREE Subscription to boards & solutions magazine 


Advertisers Index 

COMPANY PAGE 

AAEON Technology 48 

ADLINK 5 

Advantech 17 

Aitech Defense Systems 9 

Commell 45 

congatec 35 

Curtiss Wright 27 

DDC 25 

EKF 15 

Ericsson Power Modules 2 

Eurotech Ltd. 13 

Fastwel 39 

Moxa Europe 7 

MSC 3 

PEAK-System Technik 19 

Pennwell 29 

RTS 47 

TQ Components 37 

Ensure getting your personal copy of B&S magazine free of charge by completting the online form at: 

• www.embedded-control-europe.com/bs_magazine

17 th edition 

The Real-Time and Embedded Systems Show 

March 31, April 1 & 2, 2009 - Paris - Porte de Versailles 

To exhibit, visit the show or register for conferences: www.groupesolutions.com 

The 100% Embedded and Real-Time event ! 

Exhibition: 150 of the most representative actors on a unique space. 

Hardware and software solutions: ■ Processors ■ Cards and components ■ Modules 

■ RTOS ■ Environment development tools ■ Network and wireless 

Conferences-debates and workshops: 

Know-how, user testimonies and practical applications... 

Rts EMBEDDED SYSTEMS will take place along with: 

11 th Digital display 

and viewing exhibition 

and 

Exhibition strictly reserved for professionals. 

2009 4 th MtoM 

solutions show 

Organisation 97, rue du Cherche-Midi 75006 Paris - France - Tel: +33 (0)1 44 78 99 30 - Fax: +33 (0)1 44 78 99 49 - e-mail: info@salon-rts.com

AAEON’sSeriesofProductswith 

Intel® Atom Processors 

> Low Power Consumption 

>SFF(SmallFormFactor) 

> Fanless 

COM-U15 

» Onboard Intel® Atom Z530/ 

Z510 Processor 

» Intel® System Controller Hub 

US15W 

» DDRII 400/533 Memory, Max. 2GB 

» Gigabit Ethernet 

» Up to 24-bit LVDS LCD, SDVO 

» High Definition Audio Interface 

» PATASSD(Upto4GB)x1,SATAIIx1orPATAx1 

» USB2.0x8 

» PCI-Express [x1] x 1 

» Wide DC Input Range, +8.5V to +19V 

» COMExpressPin-outTypeII 

» Compact Module Size, 95mm x 95mm 

COM-945GSE 

» Onboard Intel® Atom 

N270 Processor 

» Intel®945GSE+ICH7M 



» CRT,18-bitDual-channelLVDSLCD,TV 


» PATAx1,SATAIIx2 

» USB2.0x8 

» PCIx4,PCI-Express[x1]x3 

» Wide DC Input Range, +8.5V to +19V 

» COMExpressPin-outTypeII 

» Compact Module Size, 95mm x 95mm 

NanoCOM-U15 

» Onboard Intel® Atom Z530/ 

Z510 Processor 

» Intel® System Controller Hub US15W 

» OnboardDDRII533MemoryChip,Max.1GB 


» Up to 24-bit LVDS LCD, SDVO Connector x 1 


» PATASSD(Upto4GB)x1,SATAIIx1 

» USB2.0x8 


» WideDCInputRange,+4.75Vto+14.7V 

» COMExpressPin-outTypeI 

» Nano Module Size, 84mm x 55mm 

Nürnberg, Germany 

Date: 3.-5.3.2009 

BoothNo.:Halle9Stand137 

XTX-945GSE 





» 10/100Base-TX Ethernet 

» CRT,Upto24-bitDual-channelLVDSLCD,TV,SDVO 

Connector x 1 

» AC97 2.3 Codec 2CH Audio 

» PATAx1,SATAIIx2 

» USB2.0x6 


» +5VOnlyOperation 

» XTX,114mmx95mm 

EMB-9459T 


N270 Processor (Fanless) 



» GigabitEthernetx2 

» CRT,TV-out,18-bitDual-channelLVDS,DVI,SDVO 

Connector x 1 

» AC97 2.1CH Audio S/P DIF 

» SATAIIx2,EIDEx1&CompactFlashTypeIIx1 

» USB2.0x8,COMx6,Parallelx1 

» PCIx1,MiniPCIx1,PCI-E[x1]x2inPCI-E[x4](Through 

Riser Card) 

» +12V Only Operation 

» OnboardSATAPowerConnectorx1 

» Optional TPM 

» Mini-ITX,170mmx170mm 

EPIC-9457 

» Intel® Atom N270 

Processor 




» CRT,DVI,24-bitDual-channelLVDSLCD,TV 


» USB2.0x6,COMx4,8-bitDigitalI/O 

» EIDEx1,SATAx2,CompactFlashx1 

» PCI-104 Express Expansion Connectors 

» WideDCVoltageInput:8.5V~19V 

» TouchScreenController(Optional) 

» EPICExpress,115mmx165mm 

AAEON Technology GmbH 

An der Trift 65d 

D-63303 Dreieich 

Germany 

GENE-9455 




» DDRII 400/533 Memory, 

Max. 2GB 


» CRTorDVI,18-bitDual-channelLVDSLCD,TV 


» SATAIIx2,EIDEx1&CompactFlashx1 

» USB2.0x4,COMx4,8-bitDigitalI/O 

» MiniPCIandECXProprietaryExpansion 

» +8.5Vto+19VWideDCInputRangeor+5VOnly 

Operation 

» 4/5/8-wireTouchScreenController(Optional) 

» ECXType,146mmx101.6mm 

PFM-945C 




» Onboard DDRII 400/533 

Memory Up to 1GB 

» 10/100Base-TXEthernetx1 

» CRT&Upto18-bitDual-ChannelLVDS 

» AC97 2.3 Codec Audio 

» SATAx1,CompactFlashx1 

» USB2.0x4,COMx4 

» PCI/104-Express Expansion 

» +12V Only Operation, AT/ATX Power Type 

» PCI-104-Express, 90mm x 96mm 

TEL: +4961033747900 

FAX: +4961033747949 

E-Mail:saleseurope@aaeon.com 

URL: www.aaeon.eu 

www.aaeon.eu 

AAEON’s Series of Products with Intel® Atom Processors 

AAEON is an Associate Member of the 

Intel® Embedded andCommunications 

Alliance: a communityofembedded and 

communications developers and solution 

providers.

in PDF Format - Embedded-Control-Europe.com

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?