Skip navigation


9 Posts authored by: johndonovan1

Single-purpose embedded devices typically use microcontrollers (MCUs), which are essentially small single-chip computers. However their limited capabilities and lack of flexibility make low-power, small form factor (SFF) single-board computers (SBCs) a very attractive option for applications that must work in complex environments.

But SBCs face the same constraints as smaller embedded applications: the need to be extremely energy efficient while still providing high performance. Small form factor SBCs based on Intel Intel® Atom™ processor E3800 product family (codename Bay Trail) from Intel partners provide speed and flexibility without sacrificing power or performance. This is a particularly important advantage as the number of connected, intelligent devices continues to swarm onto the so-called Internet of Things (IoT).

Intel® Atom™ Processor Powered

The new Intel Atom processor E3800 product family represents a timely, transformative response to the myriad opportunities made possible by IoT. The high-performance, low-power solution enables more intelligent devices and powers the gateways that improve data flow from device to cloud. The processors address real-world needs through excellent computing and graphics, accelerated security and image processing, an integrated memory controller with error correcting code (ECC), high throughput, and low-power I/Os that can operate over an industrial temperature range.

Atom block diagram.jpg

Figure 1: Intel® Atom™ processor E3800  block diagram

Specifically designed for intelligent systems Intel Atom processor E3800 SoCs utilize Intel’s 22 nm process technology with 3D Tri-Gate transistors. By increasing the distance between transistor source and drain the Tri-Gate design significantly reduces static power loss, while the move to 22 nm greatly reduces active power consumption; both increase computational efficiency. The new microarchitecture provides extensive power management capabilities and enhanced security. Intel® Virtualization Technology (Intel® VT) allows the operating system more direct access to the hardware, enhancing system and application performance.


The Intel Atom processor E3800 product family is available with one to four cores, 512 KB to 2 MB of L2 cache, and one channel that is configurable for Error Correction Code (ECC). Intel Atom processor E3800 product family cores operate from 1.33 MHz (E3825) to 1.91 MHz (E3845), processing graphics at up to 792 MHz (Turbo mode). The high-speed graphics capabilities enable highly efficient image processing, which is increasingly important in industrial and commercial applications.

Different Intel Atom processor E3800 product familly feature one or two memory channels and support DDR3L-1066 or DDR3L-1333 memory modules. The Intel Atom processor E3800 product family is certified for the industrial temperature range of -400C to +1100C. Overall power consumption ranges from 5W (E3815) to 10W (E3845)—very low numbers for this degree of computing capability.

Intel on Board

The Portwell WADE-8078 is a Mini-ITX embedded SBC is based on the Intel Atom processor E3800 family with memory and PCI Express controller integrated to support one-channel DDR3L memory and PCI Express 2.0 lanes. Each WADE-8078 board supports VGA, HDMI, Gigabit Ethernet, Audio, USB 3.0, SATA, and CFEX. Available in the popular Mini-ITX form factor (17 x 17 cm), these Portwell SBCs meet multiple industrial requirements for cost effectiveness, reliable performance, and a high level of data integrity and uptime.

Portwell WADE-8078.jpg

Figure 2: Portwell WADE-8978

The Axiomtek CAPA841 3.5” Embedded SBC goes all out for performance, combining a quad-core Intel® Atom™ processor E3845 running at 1.9 GHz with a 1.4 GHz dual-core Intell® Atom™ processor E3826 and up to 8 GB of DDR3L-1066/1333 SO-DIMM system memory. The CAPA851can support dual displays including full HD over HDMI. Advanced connectivity features include four serial ports, two of which are BIOS-selectable as RS-232/422/485; four high-speed USB 2.0 ports; two Gigabit Ethernet ports with Intel® Ethernet controller 12101-T; HD audio; SATA-300 port; CFast™ socket; and digital I/O. Despite its diminutive size (14.6 x 10.4 cm) the CAP851 can support two PCI Express Mini Cards (one full size and one half size).



Figure 3: Avalue ECM-BYT SBC

The Avalue ECM-BYT 3.5” SBC is available with single, dual, and quad-core Intel Atom processor E3800 SoCs. While able to drive two simultaneous displays (HDMI + VGA/LVDS) , the ECM-BYT also features an interface to 4-, 5-, and 8-wire touchscreens—or the board can be ordered with a built in touchscreen. Other I/O interfaces include one each SATA II, RS-232/422/485, and USB 3.0; 3 x RS-232; 3 x USB 2.0; and 4-bit GPI and GPO. A system temperature sensor works with an auto-throttling control to protect the system in harsh environments.

With a full range of high-speed, low-power video and connectivity options of the Intel Atom processor E3800 Soc family, the Nexcom EBC 355 3.5” SBC is ideal for battery-powered portable devices, multimedia HMI panels, outdoor systems installed in harsh environments, home automation, and thin clients. Designed for harsh outdoor environments, the EBC 355 series is a logical choice for gate control systems, gas station kiosks, and public information displays.

Powering Down—and Up

While MCUs can power a wide range of single-function embedded applications, small single-board computers provide far more power and flexibility for demanding industrial applications. The low-power SBCs discussed above can enable, for example, faster collection of patient data for portable health monitoring devices. In addition, with 3D and video hardware acceleration support, the Intel Atom processor E3800 SoC family-powered SBCs can provide faster image processing for handheld ultrasound machines and offer a more precise visualization of process controls in industrial automation.

The latest generation of Intel Atom processors—combining low-power with fast data and image processing—are making it possible for small form factor SBCs to replace older, bulkier computing platforms in a wide range of industrial applications.


Learn More

Contact Featured Alliance members:

Solutions in this blog:


Related topics:

Portwell is a Premier member and Axiomtek, Avalue, and Nexcom are Associate members of the Intel® Internet of Things (IoT) Solutions Alliance.

John Donovan

Roving Reporter (Intel Contractor), Intel® Internet of Things (IoT) Solutions Alliance

Editor/Publisher, Low-Power Design
Follow me on twitter: @jdonovan43


Collision avoidance has long been important in video games, and it’s now starting to appear in real life. High-end vehicles include a number of collision avoidance technologies that are rapidly becoming mainstream. If your current car doesn’t have them, it’s a good bet that your next car will.


Before collision avoidance technologies can help you avoid collisions, your car first has to be computerized; with as many as 100 processors in a mid-range car, that’s a done deal. The first electronic control units (ECUs) in passenger vehicles date back to 1971 when Ford introduced a 4-sensor, 3-channel anti-lock braking (ABS) systems in the Lincoln Continental. By quickly and repeatedly braking to the skid point ABS systems can reduce stopping distance by 30% compared to what a skilled driver could hope to accomplish.


Even with ABS assistance if you brake hard on a rain slick street your car can careen out of control. Electronic Stability Control (ESC) systems can detect both skidding and loss of steering control. They detect when the car is going in a different direction from where you’re steering it, then they apply braking selectively to individual wheels to keep the car on an even keel; some units may even throttle back the engine until you regain control. If a crash seems immanent the ESC system will pre-tension seatbelts. All new vehicles sold in the U.S. since 2012 have electronic stability control.


On the Radar

ABS and ESC systems help you out once a collision is imminent, but how can you foresee and forestall a probable collision? By the use of radar-based adaptive cruise control (ACC). When the radar detects a possible collision it first gives an audible warning to the driver as well as a visual signal projected onto the windshield. If the warnings aren’t heeded the ACC will initiate braking—and, if necessary, steering—in order to avoid the collision.


In the early 1970s automobile manufacturers began experimenting with millimeter-wave radar for collision avoidance, though the suitcase-size devices were hardly practical for the family car. Modern systems employ both long-range radar (LRR) operating at 77 GHz and short range radar (SRR) operating in the 24-26 GHz range.


Long-range radar is typically used to look far ahead for possible obstacles on which you’re closing; they require your attention but not immediate action. Short-range radar is better suited for crowded, urban environments. SRR can be used for blind-spot detection (BSD), raising a lane change warning (LCW) if you start to change lanes when a car is approaching from the rear. If you don’t react quickly to the warning the ESC system can prevent you from steering into that lane. Narrow-band SRR systems operating in the 21.65-26.65 GHz ISM band may use multiple antennas for beam forming, enabling them to narrow their focus to a particular approaching vehicle while ignoring nearby traffic.


Camera sensors are sometimes used in conjunction with radar in ACC systems. Backup cameras have proven their worth for city parking and just backing down your driveway. Side mounted cameras can serve as blind-spot detectors and warn of pending dangers to the side while the radars focus on the line of travel. Infrared night-vision cameras can detect pedestrians or animals in the road ahead that radar may discern but not be able to identify; using the windshield like a heads-up display the road ahead is much more comprehensible than what your headlights show on a dark night. The results from radar and camera sensors are fused with vehicle acceleration, braking, and handling systems to reduce the possibility of accidents.

Hardware Makes It Happen

One thing that all collision avoidance technologies require is low latency data processing and rapid response to pending problems. A car traveling at 70 miles per hour covers 100 feet in less than a second; a distracted driver who suddenly became aware of an obstruction in the road could easily travel the length of a football field before even applying the brakes. A radar equipped adaptive cruise control system, detecting the problem as well as the driver’s hesitation, could respond in a fraction of a second by applying the brakes and, if necessary, steering around the obstruction. Every step in this process – from detection to correction – requires a fast, flexible computing platform.


The Intel® Atom™ processor has been widely adopted in transportation applications, thanks to its speed, flexibility, low power consumption, and ability to operate in demanding environments. The dual core Intel Atom processor D2550 (formerly Cedar Trail) doubles down on these capabilities. Running at 1.86 GHz the D2550 has a memory bandwidth of 6.4 GB/s and an integrated graphics processor. Intel® Hyper-Threading Technology delivers two processing threads to each physical core, enabling highly threaded applications to get more work done in parallel, completing tasks sooner. The D2550’s Intel® 64 architecture improves performance by allowing systems to address more than 4 GB of both virtual and physical memory. In short the Intel Atom processor D2550 is particularly well suited to the applications described above.

D2550 block diagram.jpg


Figure 1. Intel® AtomTM processor

D2000/N2000 series system block diagram



The new MS-9896 Fanless 3.5” Embedded Board from Micro Star International (MSI) is a small form factor single board computer (SBC) built around the Intel Atom processor  D2550/N2800/N2600 Dual Core CPU, an Intel® GMA3650 Graphics Controller, and an Intel® NM10 Express chipset. Designed to work from a 12V supply the MS-9896 supports DDR3 1066 MHz SO-DIMM memory up to 4 GB and can drive two independent displays (VGA/HDMI/LVDS). The numerous I/O channels include 6x USB 2.0, 4x COM, 8-bit GPIO, 2x GbE LAN, and 2x PCIe.


The COM-CV Rev.B-Com Express CPU Module from Aaeon Technology is a compact COM Express Type 2 SBC. It includes a 1.86 GHz Intel® Atom™ D2550, 1.6 GHz Intel Atom processor N2600, 1.86 GHz Intel Atom processor N2800 (optional), 4 GB of DDR3 memory, 8x USB2.0, GPIO 8-bit, 3x PCI-Express, 2x 32-bit PCI00, and 1x Gigabit Ethernet. An 18/24-bit dual channel LVDS interface can support a screen resolution of 1366 x 768 pixels. The COM-CV unit can operate from 32-1400F (-40-800C) at up to 90% relative humidity.


Looking Ahead


As radar-based collision avoidance systems become more capable, the era of completely autonomous vehicles is getting closer. Adaptive cruise control systems—interacting with other vehicles and intelligent highway infrastructure—will interact smoothly with drivers, greatly reducing the frequency and seriousness of accidents.


Learn More

Contact Featured Alliance Members:


Solutions in this blog:


Related topics:


Aaeon Technology is an Associate member and Micro Star International is an Affiliate member of the Intel® Intelligent Systems Alliance.


John Donovan

Roving Reporter (Intel Contractor), Intel® Intelligent Systems Alliance

Editor/Publisher, Low-Power Design
Follow me on twitter: @jdonovan43

figure1.jpgCan your car watch out for itself in heavy traffic? The average new vehicle carries plenty of computing power, with as many as 100 processors connected by as much as a mile of wire—it’s a mobile computer network with an array of sensors controlling practically every aspect of the engine and drive train. It was only natural to start adding sensors reporting on the external environment to warn of pending danger such as merging trucks, cars approaching on your blind side, or the vehicle in front of you braking suddenly.

Automotive radar is becoming increasingly common in high-end vehicles. Such systems trigger in-vehicle warnings of imminent collisions or inadvertent lane changes and, if you don’t react quickly enough, will initiate corrective steering and/or braking. The National Highway Traffic Safety Administration (NHTSA) estimates that such sensor-based crash avoidance technologies could potentially prevent as many as 80 percent of automobile accidents involving non-impaired drivers.

Talk to Me

However, just as a PC becomes more useful when it connects to a network, vehicle-to-vehicle and vehicle-to-infrastructure communications can greatly improve both safety and mobility. The two approaches are closely tied together.

Connected vehicle systems are based on Dedicated Short Range Communications (DSRC), a two-way, short-range (approximately 200 to 300 meters) wireless communication protocol that permits secure, fast data transmission critical in communications-based, active safety applications. The Federal Communications Commission (FCC) has allocated 75 MHz of spectrum in the 5.9 GHz band for use by Intelligent Transportations Systems (ITS) vehicle safety and mobility applications. DSRC was developed with the goal of enabling technologies that support safety applications and communication between vehicle-based devices and infrastructure to reduce collisions.

The U.S. Department of Transportation is currently conducting a Safety Pilot program involving 3,000 cars with DSRC beacons that emit a basic safety message 10 times per second. This information is collected and shared with other vehicles to indicate when a potential traffic hazard exists. If a vehicle was involved in an accident it would automatically alert other vehicles as well as first responders of the incident. The accident would immediately show up on the GPS displays of properly equipped vehicles and advise their drivers of alternative routes.

As traffic built up behind the accident site, freeway signs could be activated to warn of the delay and stop lights on alternative routes retimed to handle the increased traffic. According to the Texas Transportation Institute American drivers spent 4.8 billion hours stuck in traffic in 2010, the equivalent of one full work week for everyone on the road that year—in the process wasting 3.9 billion gallons of gas.

By communicating with roadside infrastructure drivers could be alerted in advance as they approached stop lights, school zones, workers or vehicles on the side of the road, or dangerous curves. If you failed to notice a yellow light your car might automatically start decelerating at just the right rate to stop you at the intersection. In another scenario if you arrive at a stop light late at night and there are no other cars approaching, your car could signal the light to change to let you pass instead of sitting there by yourself for two minutes.

Building the Backbone


The development and deployment of a fully connected transportation system requires a robust, underlying technological platform. The platform needs to be a combination of well-defined technologies, interfaces, and processes that, combined, ensure safe, stable, interoperable, reliable system operations that minimize risk and maximize opportunities. This is pretty much a definition of the sort of flexible, compatible intelligent systems that computing platforms built around 4th generation Intel® Core™ processors (Haswell architecture) can provide.


Kontron’s CP6005(X)-SA CompactPCI Processor Boards provide an ideal backbone for powerful network intensive applications providing virtualization (VT-X, VT-D) and highest graphics performance by up to 20 graphics cores supporting OpenCL 1.2 and OpenGL 3.2 and three independent interfaces. The boards’ I/O capabilities include 10 Gigabit Ethernet, PCIe 3.0 (x4), PMC/XMC, USB, VGA, DVI, RAID, and more.


Anticipating long embedded lifecycle support, SBS Science and Technology’s COM Express™ Type 6 Module-COMe8400 provides high performance, flexibility, and X86 software compatibility. The boards include an Intel® QM87 chipset; Intel® HD Graphics with DirectX 11.1, OpenCL1.2 and OpenGL 3.2 support; 1 PCI Express x16, 7 PCI Express x1 lanes; 10/100/1000Mbps Ethernet; 2 SATA 6GB/s ports, and 2 SATA 3GB/s ports.


Venture Corporation’s eIPC380 Embedded Industrial PC is a compact computer that incorporates connectivity, manageability, and security in a ruggedized and low profile enclosure that protects the system in tough operational environments. The elPC380 I/O includes USB 3.0, HDMI, VGA, Display Port, PCIe, and SATA HDD. Wireless connectivity for 802.11 A/B/G is available via a Mini PCIe slot.


The Evoc NPC-8223 is a 2U standard rack-mount mainstream platform targeting network security applications. The NPC-8223 supports 6x GbE, 1x PCIe (x8), 4x SATA, 2x USB 2.0, and 4x 1066/1333 MHz UDIMMs (up to 32 GB). With EVOC ENM network module expansion, the NPC-82234 can support up to 14 Gigabit LAN ports.


Are We There Yet?

No, the infrastructure isn’t ready yet, but this is the direction that things are going. Intel’s wide ecosystem of partners can readily provide the architectural building blocks. Given sufficient funding and some time, driving from point A to point B will become a much safer and more efficient experience.

Learn More

Solutions in this blog:


Related topics:

Kontron is a Premier member of the Intel® Intelligent Systems Alliance. SBS, Venture, and Evoc are Associate members of the Alliance.

John Donovan

Roving Reporter (Intel Contractor), Intel® Intelligent Systems Alliance

Editor/Publisher, Low-Power Design
Follow me on twitter: @jdonovan43

smart grid.jpgComputers become far more useful once they’re networked—at which point they also become vulnerable. Despite firewalls and anti-virus software there’s hardly a PC that hasn’t been the recipient of a virus that tracks online browsing activities or sends spam to a contact list. PC viruses rarely bring down the computer, since the sender is more interested in quietly stealing the information on it or joining it to a botnet that sends out further spam and/or viruses.

The Smart Grid is essentially a large, high-voltage communications network, and as such it’s subject to hacking, just like any other network. Unlike PC viruses any attack on the grid would be disruptive and potentially catastrophic. This concern has been one of the driving forces behind the move to a decentralized, robust, secure Smart Grid.

The Smart Grid is still a work in progress, with much of the North American electrical grid still consisting of a wide range of proprietary components and protocols. They’re networked, but they were designed before cyber security became a major issue.

Get Smart

The Smart Grid is essentially a complex industrial control system (ICS), where some assets have long been part of the grid (SCADA, remote terminal units (RTUs), etc.) and others are new “smarter” assets (Advanced Metering Infrastructures (AMI), intelligent electrical devices (IEDs), smart meters, etc.). All of these are high value targets that can serve as entry points into the grid with the goal of taking over SCADA systems.

The cyber security issues are known as the “CIA triad”—Confidentiality, Integrity, and Availability:

  • Confidentiality—Access to information is largely a privacy issue; it’s important to consumers but less so for network security.
  • Integrity—Protecting the integrity of control commands is imperative in order to maintain control of the grid. This equally true in a corporate environment.
  • Availability—Continuous availability of real-time data is critical to the operation of SCADA systems, though it’s less of an issue for corporate IT systems.

The security issues for the Smart Grid are the same as those in corporate IT systems but the priorities are different:

ICT security in smart grids.jpg

Figure 2: Security issues for the Smart Grid vs. corporate IT systems

Maintaining the availability of real-time data is the top priority for Smart Grid systems, followed closely by the ability to ensure and maintain data integrity.

The power grid—with all its diverse, interconnected devices—represents an extremely large attack surface. Hardening it must start with putting all its Internet-connected elements behind secure servers with layered hardware and software security features.

Better security starts with the servers. Dell's 12th Generation PowerEdge R720t a is Tier 1 class, Network Equipment Building System (NEBS) Level-3/ETSI certified, carrier-grade server running four 95W Intel® Xeon® E5-2600-series processors. The Dell server takes advantage of the Intel® Intelligent Systems Framework, which provides a consistent way to address the foundation capabilities of connectivity, manageability and security. Rich connectivity options provide the flexibility to merge into existing deployments or legacy environments. The platform provides security, manageability and data ingestion options in addition to lightweight application functionality at a basic level.

Hardware based security features can create a trusted execution environment that prevents malicious software from running. Intel® Trusted Execution Technology (Intel® TXT) integrates security features directly into the processor, chipset, and other platform components to enable running mission-critical applications in a safe partition in hardware-secured memory regions. By storing VPN security keys and other critical data in secured memory, Intel® TXT secures the communications links along the Smart Grid.

TXT table.jpg

Figure 3: Intel® Trusted Execution Technology (Intel® TXT)

The Smart Grid relies on distributed intelligence, so the smaller computers reporting back to central SCADA servers must also be secure.

Congatec’s conga-TS87 COM Express Type 6 module is a compact, secure computing solution that can be distributed at various points along the grid. Based on the 4th Generation Intel® Core™ i7 processor the conga-TS87 includes a wide range of connectivity options including seven PCI Express Rev. 2.0 lanes, four 6 Gbps Serial ATA, 8x USB 2.0, and 4x USB 3.0. The boards can be equipped with a discrete Trusted Platform Module (TPM) that is capable of calculating efficient hash and RSA algorithms with key lengths up to 2,048 bits; the TPM also includes a real random number generator.

A Holistic Approach

Implementing cyber security on the Smart Grid is a multi-faceted problem that requires firewalls, intrusion prevention systems, event management, application whitelisting, network security design, system hardening, and security features embedded at the processor level. All of the security challenges are magnified when connecting legacy systems to new ones, which is the nature of today’s Smart Grid. Those issues can be alleviated by standardizing on a distributed computing architecture based on scalable Intel technologies that can enable the grid to be both smart and secure at the same time.


Learn More

Solutions in this blog:


Related topics:


Dell is a Premier member of the Intel® Intelligent Systems Alliance. Congatec AG is an Associate Member of the Alliance.


John Donovan

Roving Reporter (Intel Contractor), Intel® Intelligent Systems Alliance

Editor/Publisher, Low-Power Design
Follow me on twitter: @jdonovan43

Both law and logic dictate that networked transportation systems must be as secure as possible. In a previous post we explored Positive Train Control (PTC), a computerized system for monitoring and controlling the movement of trains. These are typically proprietary mission-critical wireless systems that utilize the 217-222 MHz band. They’re not open to the public, and—needless to say—not easily hacked.


There are other transportation systems—for trains, buses, heavy industrial equipment and other rolling stock—that aren’t part of a closed control loop and that utilize widely available commercial interfaces, including Wi-Fi, 3G/4G cellular, USB, Ethernet, RS232 and RS485. These are non-mission critical systems that may include passenger counting, asset management, and GPS location.


“One example is a project we’re doing for DC Metro,” explained Kurt Hochanadel, Corporate Product Marketing Manager at Eurotech. “They have a wear leveling project where we’re reporting the wear level on the wheels of the trains. We’re actually on the wheels, measuring wear and reporting to their back-end asset managers through a Wi-Fi access point technology. There are access points at depots and stations; we supply a secure, encrypted channel to the Wi-Fi and also a secure connection to the server, so it’s double encrypted [using] IPsec and VPNs. We’re using standard tools that provide the best security in the marketplace and not trying to build something from scratch.”


The heart of Eurotech’s DC Metro system is the DynaVIS 10-00 (see Figure 1), a compact, rugged mobile display computer. The DynaVIS 10-00 features a 5.7” VGA touchscreen and connectivity through Wi-Fi, 3G cellular, and Gigabit Ethernet. It’s powered by a 1.10 GHz Intel® Atom Z510PT processor with 512K cache and 400 MHz FSB and an Intel® System Controller Hub US15WPT Chipset (Intel(R) SCH US15WPT). . The device is housed in an IP65-rated enclosure and features high-end rugged connectors that provide long-term reliability in harsh environments.


Figure 1: The Eurotech DynaVIS 10-00 is an Intel® Atom processor-powered
rugged computer designed for use in the transportation industry.


The DynaVIS 10-00 is EN50155 certified—the European standard for "Railway Applications—Electronic Equipment Used On Rolling Stock”, which covers the extended operating temperature range (-25/+70 degrees Centigrade), plus resistance to the humidity, shock, vibration, and radiation encountered in vehicle or airborne installations. When asked about automotive applications, Hochanadel replied that rail requirements “are substantially more robust than for automotive. There are a lot of different applications in transportation—basically all your typical logistics. The DynaVIS 10-00 is an onboard computer that talks to all your interfaces and all your equipment.”


Secure by Design

On the software side the DynaVIS 10-00 runs Wind River Linux 3.0, from which it derives many of its security features. According to Hochanadel, “Typically most applications start from a Linux environment and meet the security requirements from that standpoint. Most of the security is done using standard IP tools utilizing SSL and SSH encryption.” Wind River Linux provides a secure and robust environment for the rest of the applications.


Security starts at the operating system level. Wind River Linux includes SELinux as a Linux Security Module (LSM), a piece of the kernel that arbitrates access to all systems resources based on security policies as well as a collection of tools for developing, debugging, and enforcing those policies. Wind River Linux also includes advanced preemptive security technologies such as run-time stack and buffer overflow protection as well as a complete intrusion detection and prevention system.


At the protocol level older Internet security systems, including the Secure Socket Layer (SSL), Transport Layer Security (TLS), and Secure Shell (SSH)—all of the DynaVIS 10-00 also supports—are implemented at the application layer of the Internet protocol suite. In contrast Internet Protocol Security (IPsec) is implemented at the Internet layer, where it can provide seamless end-to-end security between hosts and networks.

Implemented in both IPv4 and IPv6, IPsec can operate in both Transport mode—where only the payload of the IP packet is encrypted and the routing is left intact—or Tunnel mode, where the entire IP packet is encrypted and inserted into another packet with a new IP header. Tunnel mode is used to create virtual private networks (VPNs). In either case IPsec implements one of three cryptographic algorithms: HMA-SHA1, TripleDES-CBC, and AES-CBC. The probability of hacking any of these encryption algorithms when implemented with a sufficiently long key is vanishingly small.


Architected for Success

Having been designed from the beginning with Intel hardware and software, the DynaVIS 10-00 supports the Intel® Intelligent Systems Framework (ISF). Designed before ISF was introduced, Eurotech built the system said Hochanadel “with the same kind of componentry as a Java Virtual Machine (JVM) and an OSI framework to provide functionality that’s portable across different platforms.” Eurotech called this its Everywhere Software Framework (ESF). Built around Intel hardware and software and with the same design goals, it’s not surprising that the system is not just “ISF ready” but in fact ISF validated.


Asked why Eurotech chose the Intel architecture Hochanadel replied, “The hardware, software, and tools were more advanced than anything else. Also the price/performance gap favors Intel, as does its ease of wireless connectivity, especially in a Linux environment. Plus most devices have x86 drivers. There are a lot of issues that you don’t have to deal with in an x86 environment.” That’s especially true when everything is designed to work together.

Learn More

Solutions in this blog:

Related topics:

Eurotech is an Affiliate Member of the Intel® Embedded Alliance.

     Contact Eurotech>>

John Donovan
Roving Reporter (Intel® contractor), Intel® Intelligent Systems Alliance
Low-Power Design
Follow me on twitter: @jdonovan43

Recent storms and record high temperatures have put a lot of stress on the power grid. Load shifting, smart meters, and integrating alternative energy sources are all part of the solution. But how do you measure and control what’s happening at all points on a widely distributed power grid, and then how do you coordinate all these technologies in the most effective way? In short, just how smart is the Smart Grid?


Let’s start with measurement. Phasors mean one thing to Star Trek fans and quite another to utility companies.  A phasor is a complex number that represents the magnitude and phase angle of the sine waves found in electricity. By placing phasor measurement units (PMUs) at critical points around the grid utility companies can measure power quality and assess local system conditions, enabling them to respond to local disturbances (see Figure 1) as well as balance power flow over different lines and from different power sources.


Figure 1: PMU data reveal dynamic behavior as the system responds to a voltage disturbance.


For all this to work over a large grid all phasor data is synchronized to a GPS radio clock; phasor measurements that occur at the same time are called syncrophasors. PMUs take voltage and current measurements and then digitize the results using A/D converters. This data is time stamped and sent over the network to a phasor data concentrator (PDC), where it is collected and sent to a computer to be analyzed by the  a Supervisory Control And Data Acquisition (SCADA) system at a central facility.

Synchrophasors enable a flexible, efficient smart grid by utilizing distributed measurements to maximize transmission efficiency and minimize outages. However, a major obstacle to deploying synchrophasors is the lack of standards for the PMUs that communicate line conditions back to a SCADA system.


Intel, Dell, National Instruments, and OSIsoft are helping overcome this obstacle through a synchrophasor data management solution based on the Intel® Intelligent Systems Framework. This solution combines high-performance PMUs from National Instruments with a Dell 19-inch server rack. The servers run phasor PDC software from OSIsoft that collects and analyzes data from multiple PMUs (see Figure 2). This solution enables advanced visualization, analytics, and early warning systems to help utilities detect evolving disturbances and avoid widespread blackouts.


Figure 2. The synchrophasor data management solution built on Intel® Intelligent Systems framework-based platforms employs high-volume, standard computing systems used across many industries in order to reduce deployment cost and complexity.


The National Instruments PMUs can interface with sensor hardware from a wide variety of vendors and can be updated while deployed on the grid to allow for new communications protocols or more advanced analysis. They’re powered by high-performance multicore Intel® Core i7 processors.


The PMU data from the synchrophasors is delivered to centralized Dell PowerEdge servers, which provide a complete computing, networking, and storage platform with three tiers of scalable storage. The Intel® Xeon® processor-powered servers deliver data in near real time to grid operators.


Grid operators can then manage and analyze the resulting mass of data using OSIsoft’s PI System, enabling them to quickly understand and react to problems. Data security is not an issue as the PI System meets strict North American Electric Reliability Corporation (NERC) Critical Infrastructure Protection (CIP) requirements.


Key to the success of the NI, Dell, OSIsoft synchrophasor data management solution is the use of standardized, optimized, and scalable hardware and software systems—the heart of the Intel® Intelligent Systems Framework value proposition. The North American power grid is an enormously complex network based on countless proprietary legacy systems. As the network starts to add distributed intelligence, it’s critical that these systems be able to seamlessly communicate and interoperate, not just at the substation level but on a regional  and even national basis.


The Intel® Intelligent Systems Framework provides a roadmap for readily scalable connectivity, manageability, and security based on the use of standardized, well supported hardware and software architectures. With the explosive growth of data that the Smart Grid is starting to generate—and which it requires to operate—a consistent framework for building and connecting devices that interoperate over the Smart Grid is critical to its success.


The use of Intel processors all the way from the National Instruments PMUs doing data acquisition to the Dell servers, storage hardware, networking gear, and client workstations simplifies the integration, connectivity, security, and manageability of an end-to-end solution.


Validated and tested to reduce a utility’s engineering and development costs and risk, the synchrophasor data management solution uses framework-ready hardware to provide the open architecture utilities needed to precisely sync and manage transmission and distribution systems. In addition to helping utilities improve efficiency, the framework-ready components increase compatibility and speed integration. This lets utilities focus on improving energy management instead of struggling to connect, manage, and secure the hardware.


In answer to our initial question, the Smart Grid is quite smart and getting smarter all the time.


Learn More

Solutions in this blog:


Related topics:


Dell is a Premier Member of the Intel® Embedded Alliance.

     Contact Dell>>

National Instruments is an Affiliate Member of the Intel® Embedded Alliance.

     Contact National Instruments>>

John Donovan
Roving Reporter (Intel® contractor), Intel® Intelligent Systems Alliance
Low-Power Design
Follow me on twitter: @jdonovan43

If the lights in your home have ever flickered briefly during a storm—and didn’t then stay out—a lot of Smart Grid technology kicked in between the time the lights blinked out and the few hundred milliseconds later when they came back on again.


During that brief time an operation control center detected the loss of power to your area and redirected power from another substation to compensate, perhaps diverting power from a third source to compensate for the additional load on the second substation. When the break was repaired, the transformer replaced, or the breaker in your local substation automatically reset after a lightning strike, the control center automatically brought your substation back online and rebalanced the loads between all substations. In that way the Smart Grid is said to be self-healing, though humans will always have to repair line breaks and blown transformers.


Figure 1: Electrical power distribution and transmission.


There are two types of substations: primary and distribution (Figure 1). Primary substations work on the supply side, taking power from a variety of primary sources—hydroelectric, solar, wind, geothermal, and nuclear—and putting it out on the grid. This involves synchronizing highly variable inputs such as solar—which is clearly only available during the day—with wind power, which peaks at night. The substations must also regulate the loads on the power sources, which may vary considerably in capacity.


For each primary substation there may be dozens of distribution substations, which work on the demand side, ensuring load sharing between residential, industrial, and transportation end users. When a substation starts nearing its peak capacity it signals the control center to bring other sources online to get it through peak demand, avoiding the ‘rolling blackouts’ that preceded the Smart Grid.

The Smart Grid works because substations can all communicate with each of the elements under their control, sending that information back to a master control center that controls all the substations. IEC 61850 is the IEC standard for substation automation, replacing a myriad of proprietary protocols whose lack of interoperability delayed the advent of the Smart Grid.


On the Level

There are three different levels in Smart Substation architecture: the Station Level, the Bay Level, and the Process Level. Advantech provides numerous Intel-based IEC 61850 certified Smart Substation solutions in each of these areas. Its UNO-4600 series Substation Automation Computers can operate as HMI/SCADA, Terminal (serial-port) Servers, Protocol or Communication Gateways, Cyber Security Servers (UTM), and Substation/Networking Recorders.


At the Station Level the Advantech UNO-4683 provides the communication gateway between the remote control center and all the environmental monitoring and control devices at the substation; it also provides cyber security for the substation. The UNO-4683 Automation Computer is based on an Intel® Core™ i7 running at 2.0 GHz with 4 GB of DDR3 SDRAM. It provides two RS-232/422/485 isolated serial ports with automatic flow control; 2 x 10/100/1000Base-T and 4 x 10/100Base-T Ethernet ports; and six USB 2.0 ports with three domain I/O expansions.


At the Bay Level (Figure 2) the Advantech UNO-4673A protocol server provides a data gateway between intelligent devices and the station-level controller. The UNO-4673A is based on a 1.66 GHz dual-core Intel Atom processor with 2 GB of DDR2 SDRAM. Sitting on the Ethernet backbone the Advantech UNO-4672 acts as a network recorder and analyzer, passing device data back up to the station level. The UNO-4672 is powered by either an Intel® Pentium® M running at 1.4 GHz or an Intel® Celeron® M at 1.0 GHz, each with 1 GB of on-board DDR DRAM.

bay level.jpg

Figure 2: Substation automation at the Bay Level.


Finally, at the Process Level either the Advantech UNO-4671A (Intel® Atom™ D510 @ 1.66 GHz) or UNO-4673A (dual-core Intel®  Atom™D510 @ 1.66 GHz) acts as an Intelligent Electronic Device (IED) that continuously monitors the status of transformers, circuit breakers, and switch gears, warning of excessive temperature, vibration, leakage or other issues that could cause device failure.


Getting Smarter

When the lights go out they don’t just blink for everyone—sometimes they go out for hours. The basic design of the electrical power grid is over 100 years old, and it’s only gradually being computerized. Most utilities have begun to automate the restoration process by installing supervisory control and data acquisition (SCADA) systems that monitor and control line reclosers and switches, but the system is still a long way from being completely automated. Smaller cities and other customers are usually connected to their local substation by a single radial feeder. Outages to these feeders are called in by a customer to the control center, which then dispatches a person to the area to manually restore power to customers.

Implementation of automated devices such as SCADA-enabled switches and line reclosers would cut outages. Distribution circuits could also be sectionalized with SCADA-operated devices between each section. Open points that connect to other circuits could be replaced with SCADA-enabled switches. Then in the event of a failure the system could automatically isolate the problem, opening adjacent switches and rerouting power to unaffected sections by closing connections to adjacent circuits.

The Smart Grid is getting smarter, and substation automation is the key element to its success. Advantech already has a wide range of Intel-based products that can provide a complete, automated solution. It’s just a matter of time before the Smart Grid all comes together and your lineman will need to find another line of work.


Solutions in this blog:

Related topics:


Advantech is a Premier member of the Intel® Intelligent Systems Alliance.


John Donovan
Roving Reporter (Intel® contractor), Intel® Intelligent Systems Alliance
Low-Power Design
Follow me on twitter: @jdonovan43


Until recently, the United States has taken a decidedly old-fashion approach to running its rail systems. Most of the country’s signaling, switching, and train operation has been handled manually, leading to inefficiencies and unnecessary hazards. The Metrolink commuter train accident in 2008 highlighted just how dangerous manual controls could be, prompting Congress to pass the U.S. Rail Safety Improvement Act of 2008. This act mandated that approximately 73,000 miles of rail and transit infrastructure would have PTC systems in place by 2015.

PTC is a computerized system for monitoring and controlling the movement of trains. The top priority is collision avoidance, but PTC systems can also automatically regulate the speed of trains in response to rail conditions, on-board equipment malfunctions, temporary speed restrictions, and other safety considerations. While the Federal Railroad Administration envisions a National Differential Global Positioning System (NDGPS) to enable seamless train tracking and control, to date the infrastructure is still a patchwork of legacy systems and disparate approaches.


SDR vs. the Tower of Babel

One company trying to address that shortcoming is Santa Clara based Lilee Systems. Lilee’s unique software defined radio (SDR) technology and Intel®-based hardware provides complete end-to-end wireless mobility management, enabling trains moving cross country, for example, to interact intelligently with a wide range of legacy safety systems along the way.


According to Jon Adams, Lilee’s VP of Strategic Development, “There are many components to Positive Train Control. We do the onboard radios, the onboard networking processors, the wayside radios and messaging processors, and the back office mobile IP abstraction. It’s all standards-based and high security. Every train, every piece of equipment in the field—whether it’s fixed or moving—has a fixed IP address, so it becomes straightforward to manage your assets.”


Why resort to something as complex as SDR to handle what would seem to be a relatively straightforward problem? “The answer has less to do with technology than with FCC regulatory domains,” Adams explained. “If you look at the 217-222 MHz band (Figure 1), which is where much of the industry has decided it’s going to put their PTC systems, it’s under four different parts of the FCC [regulations]: it’s under Part 80, which is Maritime Mobile; it’s under Part 90, which is Business/Industrial; it’s under Part 95, which is Citizens Band; and it’s under Part 97, which is the Amateur Radio Service. You can’t operate under the Amateur Radio Service, but you can operate under the other three parts.”


Figure 1: The U.S. radio spectrum from 217-222 MHz is multi-layered.


“The challenge,” continued Adams, “is even if you build a radio that’s flexible in frequency, it still needs to meet the special requirements of whichever part in which it’s operating. But in those parts they don’t specify modulation type, data rates, coding, or other things. So having a fixed radio means you can only service one segment of a pretty small market. But a software defined radio enables you to throw a virtual switch and suddenly you’re completely compliant with Part 80 and are at 16 kbps; or you’re completely compliant with Part 90 and you’re at 9600 baud. That’s why we took the SDR approach.”


Intel Inside (and Outside) the Train

How do Lilee’s solutions leverage Intel technology? “If you look inside our Lilee Mobility Controllers—that go in the back office—or our Wayside Messaging Servers you’ll find an x86 Intel processor that’s running the whole application space. It’s a very robust architecture, and we chose it because it’s so well supported by operating systems and by the customer base. You need to look at the cost of maintaining a platform, and we felt that the Intel architecture really does help to mitigate the unknowns.”


Figure 2: Lilee's LMC-5500 Mobility Controllers provide the backbone for an integrated PTC system.


Lilee’s LMC-5500 Series Mobility Controllers (Figure 2) provide radio device management with roaming control and enable a conduit between the remote network and the back office servers. LMC-series controllers are built around the Intel® Core™2 Quad Processor Q9400 (6M Cache, 2.66 GHz, 1333 MHz FSB) and the Intel® 3210 Chipset with 82801I9B I/O Controller Hub (ICH9). SDR radios within the network establish tunnels with the LMC-5000 to allow mobile radios to move across different segments of the network without having to be aware of the underlying network topology changes.


Lilee’s Intel® AtomTM-based WMS-2000 Connectivity and Application Controllers enable back office visibility of wayside status and alarm messages, providing an interoperable gateway for PTC and legacy train control systems.


One rail system that has completely committed to Lilee’s approach to PTC is Southern California’s Metrolink. “Metrolink is the commuter heavy rail link in Southern California with 219 miles of right of way with over 200 wayside locations for signals and switches where they need to talk to a train,” concluded Adams. “Lilee’s WMS-2000 messaging server is in every one of those. These units manage all the communications from the back office network to the train. In the back office Lilee LMC-5000 mobility controllers extract the IP address so the back office can always send a message to any particular device throughout their entire system.”


While engineers will continue to drive Metrolink’s trains, PTC backup systems are in place to insure against temporary distractions ever again leading to disastrous consequences.



Solutions in this blog:


Related topics:


Lilee Systems is a general member of the Intel® Intelligent Systems Alliance. Lilee Systems is dedicated to delivering the highest quality, most reliable products and solutions for mobile connectivity across multiple market areas including railway.

John Donovan
Roving Reporter (Intel® contractor), Intel® Intelligent Systems Alliance
Low-Power Design
Follow me on twitter: @jdonovan43

Today’s cars are so complex electronically that they’re perhaps best thought of as mobile computer networks. The cars of tomorrow—which are already starting to appear today—will be increasingly connected—to the Internet, to each other, and to roadside wireless infrastructure.

The U.S. Department of Transportation (DOT) has designated IEEE 802.11p as the basis for Dedicated Short Range Communications (DS_RC), by which a vehicle can communicate with other vehicles and roadside infrastructure. DSRC enables cooperative cruise control—cruising as part of a pack on the freeway—as well as collision avoidance, electronic road pricing and toll collection, electronic parking payment, and even braking for a red light that you may not have noticed. Beyond paying for tolls and parking DSRC could turn your car into a 4-wheeled wallet, enabling you to drive through your favorite fast-food or coffee outlet without having to dig out your credit card.


In order to provide all the functionality in your car of your smart phone—including navigation, communication, multimedia, gaming, and location-based services (“Where’s the nearest Italian restaurant?”)—the average new car may have as much as a mile of wiring inside and contain over a hundred separate electronic control units (ECUs) that communicate over a variety of networks and buses. Add to that all the cool functionality that DSRC can enable and the system gets exceedingly complex.


The very complexity of in-vehicle infotainment (IVI) systems raises serious security issues, since you’re connecting systems with consumer-grade security with mission-critical systems that control the operation of the vehicle.


Getting on the bus
One weak point is the CAN bus (Figure 1), over which the various ECUs communicate. While devices on the bus may be secure, the bus is not—which means the system as a whole is not. CAN is a message-based protocol with no built-in security features.


Figure 1: The CAN bus ties together most automotive electronic control units (ECUs).


A couple of years ago the Center for Automotive Embedded Systems Security (CAESS) demonstrated the fragility of the underlying system structure. They connected a packet sniffer to the On-Board Diagnostics II (OBD-II) port to analyze CAN bus traffic. Using a wireless link they were then able to use that information to start and stop the car, race the engine, lock individual brakes, unlock the doors, and pretty much control the entire car.


Taking their hacking to the next level the CAESS team was then able to take over control of a vehicle remotely through its telematics system. They demonstrated that it’s possible to hack a car with malware inserted into an MP3 player or transmitted over a Wi-Fi connection. Devices relying on an 802.11p wireless connection may be particularly vulnerable.


Virtual IVI
While standards bodies are working on protocol vulnerability, auto makers are moving to reduce complexity by having a single ECU handle multiple functions. In these mixed-criticality systems real-time, safety-critical components must coexist with consumer infotainment applications. Developers can meet this goal with Intel® Atom™ processor-based platforms featuring Intel® Virtualization Technology (Intel® VT) and the INTEGRITY Multivisor from Green Hills Software.

“When you’re mixing consumer-grade applications and you want security, you’re always going to have maliciousness or just software that doesn’t work the way it’s supposed to,” explains Robert Redfield, Green Hills’ Director of Business Development . “That’s why you have to start at the very lowest level of software. If you’re going to have virtualization, it has to be at the microkernel level.”


Figure 2: INTEGRITY Multivisor securely partitions off guest operating systems from mission-critical applications.


INTEGRITY Multivisor is both a secure Type-1 hypervisor and an RTOS. At the heart of INTEGRITY Multivisor is a certified microkernel that provides trusted partitioning of guest operating systems, applications, and peripheral driver software (Figure 2). Multivisor supplies only a minimal set of critical services, such as process management, exception handling, and interprocess communications. Multivisor is the only code that runs in supervisor mode, while the overlying operating systems and applications run in user mode, accessing only those resources deemed appropriate by the system engineer. For example, Multivisor will prevent a guest operating system from accessing physical memory beyond what was originally allocated to the guest’s virtual machine. This prevents a stack overflow, which malware can use to take over control of a system.


To address the security issues mentioned earlier, “You would put the drivers for the CAN bus and the Wi-Fi and the cellular radio in the mission-critical part of the operating system,” continued Redfield, “where they’re under the control of Multivisor. Multivisor is built on the most highly certified real-time operating system on the planet, that is INTEGRITY. So if you put one of those communication drivers in its own partition, if something goes wrong it’s contained.”


Complete IVI platform
Mission-critical applications need to operate in near real time, which is made possible by Intel’s AtomTM processor. Intel AtomTM N2000 and D2000 processors (codename Cedar Trail) provide hardware-accelerated virtualization. Intel® Virtualization Technology (Intel® VT) speeds up the transfer of control between the hypervisor and the guest operating systems; it assists in trapping and executing certain instructions for the guest operating system, thereby accelerating performance. Intel VT is optimized for maximum virtualization performance, and its on-chip GPU accelerates 3D graphics to one or more screens while making minimal demands on the CPU.


The combination of INTEGRITY Multivisor and an Intel Atom processor provides a secure IVI platform that can run multiple guest operating systems and protected real-time applications simultaneously, using secure partitions to ensure real-time responsiveness and fault tolerance (Figure 3).


Figure 3: The combination of INTEGRITY Multivisor and an Intel Atom processor provides a secure IVI platform.



Solutions in this blog:

Related topics:


Green Hills Software is an Affiliate Member of the Intel® Intelligent Systems Alliance and plays a critical role in developing and delivering robust operating systems with virtualization and advanced development tools and embedded solutions for embedded markets such as automotive, industrial, medical, military/government, and telecommunications.

John Donovan
Roving Reporter (Intel® contractor), Intel® Intelligent Systems Alliance
Low-Power Design
Follow me on twitter: @jdonovan43

Filter Blog

By date: By tag: