1 2 3 Previous Next


525 posts

How Harry boosted the fortunes of his company by making rational choice for the embedded platform to develop ECG machines? Would you like to know, then check out my latest blog here

If you also faced some issues in your product development and then changed the embedded platform to alleviate the issues, I would love to hear from you. Please leave a comment or you can send a personal note to me

Version 3.6.6 of the wolfSSL embedded SSL/TLS library has been released and is now available for download. Release 3.6.6 of wolfSSL has bug fixes and new features including:

OpenSSH, stunnel, and lighttpd Compatibility

OpenSSH compatibility with “--enable-openssh”

stunnel compatibility with “--enable-stunnel”

lighttpd web server compatibility with “--enable-lighttpd”

SSL 3.0 is now disabled by default

We have previously announced our plans to deprecate and remove support for SSL 3.0 in the wolfSSL library, encouraged to do so by the POODLE attack. With this release, we have disabled SSL 3.0 support by default. Users who still want to use SSL 3.0 can enable it by using the “--enable-sslv3” ./configure option.

Ephemeral key cipher suites only are now supported by default

To enable static ECDH cipher suites define WOLFSSL_STATIC_DH

To enable static RSA cipher suites define WOLFSSL_STATIC_RSA

To enable static PSK cipher suites define WOLFSSL_STATIC_PSK

Added QSH (Quantum-Safe Handshake) Extension

wolfSSL, in partnership with Security Innovation, has added support for the proposed "Quantum-safe hybrid" ciphersuite. Having this cipher suite supported in the wolfSSL embedded TLS library allows two parties to use any existing ciphersuite and "quantum-safe" any traffic protected by that ciphersuite. This means that an attacker who records the traffic and later develops a quantum computer cannot go back and crack the session.

Support for QSH extension can be enabled by using the “--enable-ntru” ./configure option.

SRP is now part of wolfCrypt

SRP is a password authentication and key-exchange protocol suitable for authenticating users and exchanging keys over an untrusted network designed by Thomas Wu at the Computer Science Department of Stanford University.

Support for SRP in wolfCrypt can be enabled with the “--enable-srp” ./configure option.

Certificate handshake message fragmentation support

Certificate handshake messages can now be sent fragmented if the record size is smaller than the total message size, no user action required.

DTLS duplicate message fixes

Visual Studio project files now support DLL and static builds for 32/64bit

For information on compiling wolfSSL with Visual Studio, reference Chapter 2 of the wolfSSL Manual, or the “Using wolfSSL with Visual Studio” webpage.

Support for new Freesacle I/O

Freescale KSDK and Kinetis Design Studio users can now compile wolfSSL for the new KSDK version of MQX by defining FREESCALE_KSDK_MQX in settings.h or by adding it to the list of preprocessor defines.

FreeRTOS FIPS support

This release includes FIPS support for FreeRTOS platforms.

This release contains no high level security fixes that requires an update though we always recommend updating to the latest version of wolfSSL.

For more information about using and compiling wolfSSL, please visit the wolfSSL Documentation page or wolfSSL Manual. If you have questions about the wolfSSL embedded SSL/TLS library, or about using it in your project, please Contact Us.

Download wolfSSL 3.6.6: https://www.wolfssl.com/

The term “prana” refers to the cosmic life force that originates in the sun and enters the body as we inhale. In the act of breathing, we form a connection between ourselves and the universe. To better explore the relationship between both breath and light, B-Reel Creative developed an installation of 13,221 LEDs that gave onlookers a visual representation of this connection.


The creative firm’s exhibit PRANA was comprised of a 12’ x 12’ sphere suspended from the ceiling, which visitors were prompted to step into and stand before a XeThru respiration sensor (ATSAM4E16E) that detected their breath. With every inhale and exhale, the XeThru data was fed into a custom JavaScript library, triggering color shifts and animations to make it appear as if the installation was breathing with them.


Sounds by One Thousand Birds were emitted to enhance its effect, seamlessly transitioning between each phase of the experience. The team open sourced the code to enable other artists and developers to create custom animations that can later be incorporated into the piece.

PRANA was designed and built entirely in-house over the course of a year and had been on display at the Fridman Gallery in New York City up until earlier this month. However, you can see it for yourself below!

This blog originally appeared on Atmel Bits & Pieces.

A few months ago, Norwegian sensor developer Novelda unveiled a pair of adaptive Atmel | SMART ATSAM4E16E powered sensor modules capable of monitoring human presence, respiration and other vital information in real-time. Based on the company’s XeThru technology, they use radio waves rather than infrared, ultrasound or light, enabling them to ‘see through’ an assortment of objects ranging from lightweight building materials to blankets. These modules can even be employed to detect movement in a room, as well as measure the breath of a person, without contact.



With one of these boards on hand, Maker Øyvind Dahl decided to build an Arduino radar that could control the brightness of an LED with his breath. To accomplish this, he hooked the X2M200 respiration monitoring sensor up to an Arduino Uno (ATmega328). The module, which was tasked with detecting his chest movement, also required both a USB communication board and a level shifter to interface the 2.8V levels of the XeThru with the 5V of the Uno.

Dahl connected an RGB LED to the project, whose brightness faded in unison with his breathing. When he inhales, the light fades in. And when he exhales, it fades out. Beyond that, the faster that he breathes, the faster the LED will fade.



In order to make this work, the Maker wrote his own code for the radar — which can be found on GitHub here — that would would send over the respiration data.

“Another thing I was struggling with, was type conversion. And with only an RGB LED as my output, it was a bit hard to debug,” he explains. “So I connected another Arduino with SPI, that I could use as a debugging console for a while, and got the type conversion sorted out. I parsed the data that was coming in, and used the movement-value to set the brightness of the LED.”


Since the Arduino did not process the data quick enough, Dahl ended up adding a function to his code that would empty the buffer and sync the data each time that he fetched a new measurement.

While this is merely a prototype, there’s plenty of potential for development. Dahl says that upon completion of his “useful device,” he will share the code and detailed breakdown of the project on his site. As we await to see what the Maker comes up with, you can find his first tutorial here and see it in action below!

This blog originally appeared on Atmel Bits & Pieces.

Do you recall what you were doing back in the summer of fourth grade? Chances are you weren’t creating a programmable, Arduino-based smartwatch like eight-year-old Maker Omkar Govil-Nair, let alone starting your own business.


The O Watch is built around an Arduino Zero (Atmel | SMART SAM D21) module and packs quite the punch when it comes to portable computing. Not only can it tell time, the wearable device can run a wide range of games and applications. For instance, the smartwatch can calculate the value of Pi, play a recognizable version of “Flappy Bird,” “Pac-Man” and “Rock, Paper, Scissors,” and collect measurements in science experiments, among many other things.

Now live on Kickstarter, the O Watch will come in two different models: a base kit and a smartwatch kit. The first is comprised of a SAMD21G18A based programmer board along with a mini color OLED screen, a LiPo battery, a 3D-printed case, and a paracord available in four colors (orange, yellow, pink and blue). Meanwhile, the latter features all of that plus a sensor board equipped with a Honeywell three-axis compass, a Silicon Labs temperature and humidity sensor, and a Bosch barometric pressure sensor.


To bring his idea to life, Govil-Nair has partnered with TinyCircuits for the design and manufacture of the watch’s electronic parts. Helping to reduce the gadget’s inner workings and thickness, TinyCircuits developed a new Arduino module with a color OLED screen, microUSB programmer and charger, all rolled into one board. The O Watch is driven by the highly-popular TinyDuino platform, while its integrated microUSB port is used for both charging and uploading programs.

“Since it’s a fully Arduino-compatible product in a tiny package, you can do a lot more – pretty much anything that is possible using a regular Arduino board and a color screen,” he explains.


What’s truly exciting about this project is that its programmability opens the door for young Makers to explore their imagination, enabling Arduino lovers of all ages to devise games and apps of their own that can be worn around the wrist.

Interestingly enough, Govil-Nair was inspired “to make his own product” after meeting our good friend and fellow whiz-kid Quin Etnyre at Maker Faire two years ago. And it looks like he’s well on his way to following in the footsteps of Etnyre with a successful crowdfunding campaign. The O Watch is currently seeking $15,000 on Kickstarter and expected to begin shipping in February 2016.

We can’t wait to see the wearable on display at the World Maker Faire in New York next month. Until then, ‘watch' it in action below!

This blog originally appeared on Atmel Bits & Pieces.

Atmel has expanded its popular lineup of secure, ARM Cortex M0+-based MCUs with the new SAM L22 series. The Atmel | SMART SAM L family is the highest scoring product family in the EEMBC ULPBench and offers an ultra-low power capacitive touch with a segment LCD controller that can deliver up to 320 segments, making the devices ideal for low-power applications such as thermostats, electric/gas/water meters, home control, medical and access systems.



The Internet of Things is driving connectivity in various battery-powered devices making security and ultra-low power critical features in these devices. With this in mind, the SAM L22 series boasts 256-bit AES encryption, cyclic redundancy check (CRC), a true random number generator, Flash protection and tamper detection to ensure information is securely stored, delivered and accessible. To get the lowest possible power consumption, the devices use Atmel’s proprietary picoPower technologies and smart low-power peripherals that work independently of the CPU in sleep modes. The latest MCU can run down to 39µA/MHz in active mode, consuming only 490nA with RTC in backup-mode.

“As more devices in the consumer, industrial and home automation segments are becoming smarter and connected, these devices require a number of unique features including ultra-low power, security, touch capability with an LCD — all features that are currently provided in the SAM L22,” explained Oyvind Strom, Atmel Senior Director of MCUs. “Atmel is already engaged with a number of alpha customers developing metering, thermostat and industrial automation solutions based on the new Atmel | SMART SAM L22 series.”

In addition to segment LCD supporting up to eight communication lines, capacitive touch sensing and built-in security measures, the SAM L22 includes up to 256KB of Flash and 32KB of SRAM, crystal-less USB device, programmable Serial Communication modules (SERCOM) and Atmel’s patented Event System and Sleepwalking technologies.

Those wishing to accelerate their designs will be happy to learn that the new SAM L22 Atmel Xplained Pro is now available. This professional evaluation board with an on-board debugger and standardized extension connectors is also fully supported by Atmel Studio. While the Atmel SAM L22 series is currently sampling, production release is slated for December 2015.

This blog originally appeared on Atmel Bits & Pieces.

It is a lonnng summer with most (little) activity coming from Atmel. Actually, we do have great news!

Keremi expands its scope and will track... Read on

A quick review of Ethernet switches and transceivers will highlight their importance in the automotive Ethernet and the OPEN Alliance.

By John Blyler, Editorial Director, IoT Embedded Systems

The connectivity that we’ve long enjoyed via the Internet at home and in the office is quickly moving to our automobiles. By 2025, the mobile operators group GSMA predicts that every car will be connected to the Internet. One of the easiest and most cost effective way to implement in-vehicle networking is via the tested and proven ethernet

In fact, there is a consortium whose sole aim is to make consumer-level Ethernet capable of meeting the automotive industry’s stringent requirements. The OPEN (One-Pair Ether-Net) Alliance Special Interest Group (SIG) is a special interest group formed by BMW, Broadcom, Freescale, Harman, Hyundai and NXP to establish BroadR-ReachTM as an open standard for automotive connectivity. BroadR-Reach technology is an Ethernet physical layer standard that allows multiple in-vehicle systems to simultaneously access information over unshielded single twisted pair cable.

As in the home and office, switches and transceivers will form the backbone of the automotive Ethernet. Let’s have a quick review on both these devices.

What are Switches and Transceivers?

Switches connect multiple devices – usually computers – together within the same network. They are generally used to expand the capacity of a network by adding more ports. Switches are designed to pass traffic through as quickly as possible, typically to a router which acts as “the brain” of the network.

In the home and small office, an Ethernet switch is typically part of the router. The latest switches support gigabit speeds.

The best switches use Deterministic Ethernet technology to guarantee message latency in applications such as autonomous driving. Deterministic Ethernet refers to a networked communication technology that uses time scheduling to bring deterministic real-time communication to standard IEEE 802 Ethernet. This technology effectively permits time-scheduled traffic to be partitioned from all other network traffic and is therefore immune from disturbance. This means that in a Deterministic Ethernet network, latency of critical scheduled communication is guaranteed.

Further, Deterministic Ethernet supports the trend toward increasing bandwidth requirements of up to one gigabit, while ensuring high reliability in networked control systems and high availability in fail-operational applications.

A transceiver - short for transmitter-receiver – is a device that both transmits and receives analog or digital signals. In a local-area networks (LANs), the transceiver actually applies signals onto the network wire and detects signals passing through the wire. This is all done on the physical layer or PHY. In Ethernet networks, a transceiver is also called a Medium Access Unit (MAU).

For automotive applications, it’s important the transceiver supports low power modes. When the engine is off, the systems must sleep. Meanwhile, the Ethernet PHY stays partially powered, waking up the system only upon activity on the network. Modern Ethernet PHY designs do not require additional components like voltage regulators to stay on while the engine is off, which greatly improves power consumption and battery lifetime.

Figure: Switches (green boxes) and transceivers (orange box) for the backbone of the automotive Ethernet platform. (Courtesy of NXP).

Data and Connectivity

The adoption of Ethernet is quickly accelerating with the rise of secure connected cars and the subsequent high demands for data transport.

“Automakers plan to use in-vehicle Ethernet broadly to enable a variety of applications and functions,” said Thilo Koslowski, vice president and automotive practice leader at Gartner. “These include safety, driver information systems, advanced driver assistance systems (ADASs) and entertainment. By 2023, 162 million Ethernet nodes containing 242 million ports will be included  in produced consumer vehicles, worldwide.”

All of the members of the OPEN Alliance and many more are supporting the adoption of Ethernet technology into tomorrow’s automobile. For example, NXP has shipped six billion CAN, LIN, and FlexRay transceivers to the global automotive industry and two million transceivers are supplied every day. Ethernet will complement these existing standards.

Recently, NXP announced its new OPEN Alliance BroadR-Reach™ product portfolio for the automotive Ethernet: the Ethernet swtich (SJA1105) and low-power transceiver (TJA1100). For more information, check out the follow links:


Originally posted on Linkedin Pulse for John Blyler

One of the more popular CPUs, if you can call it that, that's based on the ARM architecture, is the Zynq, developed by Xilinx I phrase it that way because the Zynq is not a traditional microprocessor. It's actually an FPGA that employs an ARM core.



To that end, Micrium has ported its operating-system kernels to the latest Zynq platform, known as the UltraScale+ MPSoC, a heterogeneous multi-processing system-on-chip (SoC). According to Micrium, its OS is "the only commercial RTOS that lets customers select a single OS for all of their embedded systems needs on the chip."


The combination of scalability and functionality that are offered by the µC/OS-II and µC/OS-III OSs and the FPGA suit them for a wide range of applications including embedded vision, ADAS, industrial Internet of Things, and 5G wireless systems. These are applications where reliability, security and connectivity are critical.


The Micrium repository for the Xilinx SDK, including µC/OS-II and µC/OS-III and components, including µC/TCP-IP, µC/USB, µC/FS, (among others), for the Zynq Ultrascale+ can be downloaded from the Micrium download center. As part of the company's Evergreen program, the MCU, tool chain, and embedded software components are always compatible and use the latest versions available.

Rich Nass

The Makers need love, too

Posted by Rich Nass Aug 6, 2015

The ARM architecture has certainly made a dent in the Maker movement. If you look at the high-volume development boards that are very much in the spotlight—BeagleBone - Texas Instruments, Arduino, Raspberry Pi, and so on—they're all built around an ARM core.


One of the key reasons to employ one of these boards is that a lot fo the work that you have to do is already done for you. That includes things like drivers and other code, and in some instances, it's well documented. And the manufacturers are doing their part in this initiative. For example, Atmel has about 150 people in its tools group that are designing products to make the process easier for makers (and of course for their high-volume customers, too).


From the Makers perspective, they generally go for something that's not leading edge, for a couple of reasons. One, the longer you wait, within reason, the more time there is for other people to make open-source software available for you. And tow, the prices come down over time, so the boards are very inexpensive, in many cases under $50, and in cases, under $20.


The on-going debate is whether the bigger manufacturers should concern themselves with the Maker crowd. The majority of these designs either go nowhere or end up in quantities of ten or less. It's hard to make a business selling semiconductors in those quantities. However, there's always that one (hopefully more than one) that hits it big.


According to Andreas Eieland, director of product marketing for Atmel's MCU group, "We don't know what the next big thing is going to be. There are people out there that are a lot smarter and more visionary than us, so we support all of them. In terms of tech support, we only have two tiers. There's Tier 1 and then everybody else. If you have a tech support question for us, we don't ask what volumes you plan to ship."

This article http://enable.atmel.com/cortex_M7 provides a short description on how the SRAM architecture in Atmel's ARM Cortex-M7 based MCUs releases the performance of the ARM Cortex-M7 CPU to your application.

Back in May, former Pixar developer Erin Tomson unveiled a new set of plug-and-play boards designed to take the headache and hassle out of building electronics. Not long after its Kickstarter launch, Modulo flew by its $10,000 pledge goal having garnered over $50,000 from 315-plus backers. Since then, the Richmond, California-based startup has experienced tremendous popularity at Maker Faires and has even demonstrated its simplicity with some DIY projects of their own, ranging from a tea-brewing robot to a smart sous vide machine.


Essentially, Modulo is a series pre-made circuit boards that provides Makers with all of the necessary tools to bring their gizmos and gadgets to life, without the messiness of wiring and soldering. Each module is equipped with its own little processor (ATtiny841) that is tasked with handling its operation and communicating with a controller board. While the Modulo Controller had been built around the mighty ATmega32U4 for its crowdfunding debut, Tomson has since upgraded its design to include the much faster and powerful Atmel | SMART SAM D21 — the same Cortex-M0+ MCU at the heart of the Arduino Zero. What this means is that the Controller will work nicely with Arduino and will be well received by the flourishing DIY community.

“This new chip is four times faster, has eight times the Flash storage, and has 12 times the RAM of the ATmega32u4 used in earlier prototypes,” Tomson explains.


Using a connector on its back, Makers can easily slide their boards right into the so-called Modulo Base which securely holds them in place. Following a successful Kickstarter run, Tomson had decided to also switch the connectors, both for attaching each Modulo to the base and for cables that link the bases together. These improved connectors are easier to assemble and more compact. Furthermore, those wishing to employ a Spark Core, Photo or Electron instead of the Controller can do so by selecting a Spark Base.

The Arduino-compatible Controller boasts six I/O ports that can be used as digital or analog inputs and digital outputs. Four of the six ports can even be used to control servos or output a PWM signal. Additionally, each port has its own power and ground pins to help keep things nice and neat, while circuitry on the board will protect it from any potential wiring mishaps.


Similar to other DIY dev kits like littleBits, Modulo features a number of different modules with varying capabilities. These include a color OLED display, a push-button illuminated knob, a motor driver, a thumb joystick, a temperature probe, I/O and extension cables, as well as an IR transceiver and a Blank slate that lets Makers devise circuits from scratch. Any four modules can be connected to the Base, or can be daisy chained together for larger projects.

The ARM Cortex-M0+ driven Controller can also act as a bridge, enabling users to manage their modules from Python running Raspberry Pi or a Mac, Windows or Linux computer. Beyond that, they can choose to use the Arduino IDE to reprogram the Controller or connect to the Internet via Spark. Communication between devices is accomplished through the standard I2C bus.


The Modulo Protocol allows for the Controller to dynamically discover connected devices, assign addresses, retrieve device capabilities and detect bus errors. It is an extension of I2C and can be utilized on a mixed protocol bus along with SMBus and traditional I2C devices.

“Modulo wouldn’t have been possible without the contributions of the open source community, so we’re giving back by open sourcing our protocols, hardware designs, firmware and libraries,” Tomson adds.

Those wishing to learn more, explore technical specs or pre-order a set of Modulo can head over to its official website here.

This blog originally appeared on Atmel Bits & Pieces.

It is a very simplistic view, but, broadly speaking, there are two ways to design an embedded software application [or pretty much anything really]: either top-down or bottom-up. There are pros and cons of each approach and it is hard to say, in the general sense, that one is better than the other.

So much for design, but how development is approached is another matter …

Software design is not a topic that I talk about much. This is not because I have no experience in that area – I have been designing software all my adult life. It is simply that, in recent years, I have worked for a company offering a number of tools and IP for embedded software development, but not design. And they pay the bills. However I think it is safe to say that most people tend to favor top-down design. First define the overall functionality of the application, then break it down iteratively into simpler and simpler functional units.

To read the rest of this entry, visit the Colin Walls blog.

C     yet more Questions and Answers   Mentor Graphics.png

Rich Nass

ARM comes to PC/104

Posted by Rich Nass Aug 3, 2015

The ARM architecture continues to make inroads into the embedded space. A good example of that is an announcement by Sundance, a supplier and manufacturer of PC/104 embedded computing modules. Taking advantage of the latest Xilinx Zynq SoC, which integrates dual-core ARM Cortex-A9 CPUs, the EMC2-Z7015 incorporates four lanes of PCI Express (PCIe) and re-programmable logic with Xilinx's Artix-7 FPGA technology. The module is designed it into the latest PC/104 form-factor, known as OneBank, which offers the ability to stack the modules.


ARM blog 9, photo 1.jpg


The module's ARM core is clocked at up to 1 GHz and is supported by 1 GByte of DDR3 memory, an SD-Card interface for standalone booting/local storage, USB 2.0, HDMI, SATA, and 1-Gbit/s Ethernet interface. The EMC2-Z7015 can run standard Linux applications and is supported by Xilinx's free Vivado 2015.2 tools. The tools can provide the VHDL synthesis for FPGA programing of the logic and control, where necessary.


The EMC2 -Z7015 is a single-board-computer (SBC) by default, but the real benefit comes from the stackability, where multiple SBCs can be integrated into a multiprocessing ARM system, using PCI Express as the interconnect between each Zynq SoC. The module introduces what the company calls a "cableless breakout" concept, which removes the requirement for cabling from the PC/104 board to the external world. This is implemented using Samtec's Razor Beam self-mating connector.

Inforce Computing will be exhibiting at the ‘Robotics as a Business’ (RaaB), conference organized by the Silicon Valley Innovation Center (SVIC) in San Mateo, CAhttps_proxy (962×426)On day-2 of the conference (Friday, July 31st, 2015), SVIC will host a select few companies, including Inforce Computing, in the exhibit area. Come visit us at the Silicon Valley Innovation Center between 10:00AM and 4:00PM to see two interesting robots performing cool tasks powered by Inforce’s ARM-based embedded computing platforms, such as the Inforce 6410Plus SBC. The open-source 3D-printed Snapdragon Micro-Rover running Android and the TurtleBot running ROS on Linaro Ubuntu Linux, are exciting platforms to start prototyping robotics projects. Also on display will be the latest 4K Ultra HD video capture capable Inforce 6501 Micro SOM, which is ideally suited for “SWaP” constrained commercial drones and UAVs in agriculture, surveillance, law enforcement, and other industrial applications.



From the Inforce Computing Desk
©2015, Inforce Computing, Inc. All rights reserved.

Filter Blog

By date:
By tag: