Intel’s New Fanless Apollo based NUC mini-PC and SBC released

Specs for Intel’s NUC 8 rugged mini PC and 3.5-inch dashboard have been released. The fanless NUC 8, code-named Chaco Canyon, runs Windows or Linux on low-power (6W TDP) dual-core and 1.1 GHz/2.4GHz Celeron N3350 from Apollo Lake with dual HDMI ports, 4GB RAM, soldered down 64GB eMMC, and M.2 22 x 80 key M slot for PCIe x4 NVMe.

The recently revealed mini PC designed for greater reliability has a ventless design for minimizing particle intrusion, a wide-range DC input (12-24V) with transient voltage suppression, compliance with EMC/RF standards, a delayed AC start and DC over-voltage protection. It offers ACPI power management and hardware monitoring. According to Intel, the PC was designed to fit for round-the-clock tasks “…the Intel NUC 8 Rugged is built to last and qualified for 24×7 sustained operations – a critical feature for IoT and commercial operations. Fanless and ventless, the Intel NUC 8 Rugged isn’t afraid to get a little dirty”.

The system is equipped with dual HDMI ports – HDMI 2.0a port (that supports up to 3840 x 2160 pixel display resolution, 7.1 multi-channel digital audio) and HDMI 1.4 port with CEC and EDID. It has an internal 4-lane eDP display connector, an analog audio jack, multiple mounting options including VESA mounting holes, zip-tie indentations, and 3rd party DIN-rail bracket. It has support for eight USB connections with dual rear USB 2.0 ports, front and back USB 3.0 ports and additional headers (1 x 3.0, 2 x 2.0). It also has support for a Kensington lock with base security, an internal RS-232 interface and a front-panel header. There is an M.2-based Intel Dual Band Wireless AC 3168 module, GbE ports, internal antennas, Bluetooth 4.2, Intel Gigabit LAN for wireless connectivity, and available punch-outs for adding external antennas.

The products are aimed at applications in edge analytics, quick service restaurants, hospitals, medical clinics for mobile pedestrian computing, operator consoles in manufacturing environments, digital signage and IoT. The mini PC, as well as the NUC8CCHB baseboard-based SKU with a 3.5-inch form factor, is due to ship by the end of the month.

The full kit is available for pre-sale at various locations for $248 plus $10 shipping but pricing or availability info has not been listed yet for the NUC8CCHB baseboard model. However, the 146 x 102 mm baseboard model adds a 19VDC, 65W power adapter and like the mini PC kit, offers a warranty of 3 years. For full specs check out the NUC 8 Rugged product brief (PDF) and the Ark page.

Top 5 IoT Cloud Platforms exist today

The internet as we knew it, a few years back, was being run by humans; the majority of everything that happens over the internet – the messages, the data, the communication was between people. But this is changing gradually; a new category of devices that do not have any communication between humans is beginning to take over the internet. These machines talk to other machines and are just simply being referred to as “Things”.

However, as the need to connect these devices arises, there becomes an increasing demand for a place to store, process and send information within these devices.  With millions of devices connected to the internet, we see an increasing potential of tapping and processing efficiently the data acquired from them.

Cloud computing has become a generally appreciated way of running and developing tech solutions especially the IoT which depends directly on the internet. Cloud platform for IoT brings together the capabilities of cloud platform and IoT devices as service over an end-to-end platform.

There are already plenty of cloud services available for the Internet of Things, many tech companies now capitalize on providing options for deploying IoT applications on the go. Here is a list of the top 5 IoT  cloud platforms that exist today:

1. Amazon Web Services IoT platform

No matter the kind of cloud-based project you have in mind, the AWS platform guarantees almost a 100% probability of supporting it. Because they have put in a lot of effort into innovation and building features over the years, the AWS platform probably has the most comprehensive set of tools available. The platform supports HTTP, WebSockets and MQTT. Amazon Web Services owned by Amazon allows users to host and manage services on the internet. It gives users much ease with collecting data from internet-connected devices.

The AWS offers a lot of benefits for users of this platform which includes database management, infrastructure management, cloud base data storage solutions, and application transfer. The major features of t his platform are:

  • Authentification and encryption
  • Device management
  • Device shadow
  • Secure gateway for devices

It is open-source and flexible and has good integration with IAAS offering but the AWS platform is not secure for hosting critical enterprise applications and there is a heavy learning curve for AWS. Pricing, however, is a bit on the high side here.

2. Microsoft Azure IoT hub

Microsoft is another giant in the IoT space taking cloud services very seriously. The Microsoft Azure has multiple services for IoT solutions that enhance productivity and profitability with pre-built connected solutions. It can also easily analyze untapped data or act on new data to transform business. It supports HTTP, MQ Telemetry Transport, and Advanced Message Queuing Protocol. Its features include:

  • Rich integration with SAP
  • Dashboards and visualizations
  • Easy device registry
  • Real-time streaming

While the Azure is secure, scalable and highly available, it does not have supports for bugs and also requires heavy management.

3. IBM Watson IoT Platform

IBM Watson is another giant for IoT cloud service. They provide easy apps and interfaces that make their cloud services as accessible as possible to beginners. The IBM platform has a secure communication connection, cognitive systems and offers a real-time data exchange. Recently, it added data sensor and weather data service to what to improve its customers’ service. Again, while it can handle huge quantities of data, it takes a lot of time for integration. It also has a high switching cost and needs a lot of maintenance.

4. Google Cloud Platform

Google’s IoT cloud platform is also one of the best platforms we currently have with a focus on making things fast and easy. It has an end-to-end- platform for IoT services to help scale your business. You can take advantage of Google’s heritage of machine intelligence, analytics, and web-scale processing to build IoT initiative.

Google provides efficient and scalable services, has huge storage capacity for data and cuts costs for server maintenance. It also has the ability to analyze big data and create business through a fully responsive, intelligent and protected IoT data. It takes lesser access time and gives the fastest input/output. However, the cons of this platform are that it has a limited programming language and most of its components are Google technologies.

Pricing is usually cheaper here than in Amazon and a few other platforms.

5. Oracle

This seems to be focusing on logistics and manufacturing operations but they also help to get your products to the market faster. They offer endpoint management, real-time IoT data analysis, and high speed messaging where for real-time message notifications. Its service to users includes an integrated secure and scalable service with real-time insight.

Meanwhile, like every IT system or service, each solution comes with its own advantages and disadvantages. When trying to choose the best platform for your IoT project, the end-to-end requirement and cost-benefit analysis between open-source and commercial solutions need to be considered.

Comprehensive Hardware Codec solution using iWave’s Zynq UltraScale+ FPGA SOM

Zynq UltraScale+ MPSoC / FPGA System On Module

Over the past years, live video streaming has undergone major technological advancements, triggering large scale implications in various commercial and industrial applications such as security surveillance system and online video broadcast. Advancements in video compression standards such as H.264 (AVC) and H.265 (HEVC) have made it possible for users to stream Full HD and 4K videos in real-time with high quality and detail, while taking up less bandwidth and processing time.

An application-specific processor does not provide the necessary compute resources to implement real-time encoding or multi-stream encode/decode at the highest level.

Implementing proprietary algorithms in programmable logic provides the flexibility and future enhancement to address the emerging market needs. These types of implementations typically have a programmable processor to handle application-level tasks, video codecs and manage the flow of data to the programmable logic devices.

Zynq UltraScale+ MPSoC is an ideal hardware platform to overcome all the challenges faced by the ASIC platforms. Unlike typical application processors, Zynq UltraScale+ MPSoC integrates heterogeneous high-performance ARM® multicore Cortex A53 CPU, Real-Time Cortex R5 CPU, Graphics Processor, Video Codecs, multiprocessing system with programmable logic cells. Quad-Core application processor equipped devices to deliver maximum scalability and are capable of offloading critical applications such as graphics and video pipelining, to dedicated processing HW codecs, along with a full complement of integrated peripherals, connectivity and a space for the programmability in FPGA.

The MPSoC supports the following various hardware codec features;

  • Simultaneous Encode and Decode through separate cores
  • H.264 high profile level 5.2 (4K@60fps)
  • H.265 (HEVC) main, main10 profile, level 5.1, high Tier, up to 4K@60fps
  • 8 and 10-bit encoding
  • 4:2:0 and 4:2:2 Chroma sampling
  • Low Latency Mode
Zynq Ultrascale+ MPSoC development Kit

SOM Solution offered by iWave:

iWave Systems have devised a SOM solution for implementing these advanced hardware Codecs using the latest FPGA integrated SoC technology which offers a lot of flexibility to the user by having the high-end ARM cores, graphic cores together with varying programmable logic density.

SOC: Xilinx Zynq UltraScale+ ZU4/ZU5/ZU7

  • Quad/Dual Cortex A53 @ 1.5GHz
  • Dual Cortex R5, ARM Mali 400 MP2
  • 265, H.264 Video Codecs
  • ZU4, ZU5, ZU7 (CG/EG/EV) devices
  • Up to 504K Logic Cells
  • 16 x FPGA transceivers @16.3Gbps
  • 4 x PS transceivers @6Gbps
  • 24 LVDS Pairs/48 SE IOs + 48 SE IOs

Memory:

  • 2GB/4GB DDR4 for PS with ECC (Expandable)
  • 1GB DDR4 for FPGA (Expandable)
  • 8GB eMMC Flash
  • Form Factor: 95mm x 75mm
  • BSP Support: Linux 4.14.0
  • Temperature: Industrial Grade Operation

iWave Systems has the stock of various ZU4, ZU5, ZU7 MPSoC based SOMs and also provides the comprehensive development board for immediate evaluation.

For further information or enquiries please write to mktg@iwavesystems.com or contact our Regional Partners.

Thermal-packaged MOSFET targets e-bike power systems

Analog and mixed-signal semiconductor company MagnaChip Semiconductor (Cheongju-si, South Korea) has announced the release of a 100-V mid-voltage MOSFET with a new thermal package designed for the fast-growing electric bike (e-bike) market.

The MOSFET is housed in a M2PAK-7P package suitable to meet the particular requirements of electric bike (e-bike) systems, which are powered by electric motors running on lithium-ion batteries. The 100V MOSFET, says the company, features a best-in-class thermal package and is an essential electronic device that controls the speed of the motor and supplies a stable current of electric power to the lithium-ion Battery Management System (BMS).

“The emergence of e-bike is a highly attractive opportunity for MagnaChip,” says YJ Kim, CEO of MagnaChip. “Our 100V MOSFET product in a specially-designed thermal package is designed for the particular needs of motors and lithium-ion batteries that power e-bikes like electric scooters. MagnaChip plans to launch new Mid-Voltage MOSFET products with M2PAK-7P to meet the evolving needs of this rapidly-growing consumer sector.”

The company’s 100V mid-voltage MOSFET is said to be especially suitable for e-Bikes requiring high power systems. With its M2PAK-7P package, says the company, the new MOSFET achieves a best-in-class thermal performance and a lower Rds(on) compared to other MOSFET products currently on the market.

The device also increases the maximum operating current to 240 A. These features enable a high level of operational efficiency and a secure supply of power, both of which are necessary for high-powered e-bikes.

The company says it also plans to introduce 80- and 135-V mid-voltage MOSFET products in the M2PAK-7P package. With this strategy, says the company, it plans to aggressively expand its presence within the e-bike market and become an industry leader for all power products within it.

MagnaChip Semiconductor

Using Long Range 315MHz RF Wireless Transceivers with Arduino

Connectivity continues to be one of the most important features of any device in recent times, even for devices that are not directly connected to the internet, the need to send data from one device to the other is an important part of the ubiquitous world being built today. For designers, DIY hobbyist, and professionals alike, the choice of selecting the right communication module depends on the knowledge of the available options, as such, for today’s tutorial, we will look at the use of Long Range 315MHz RF Wireless Transceivers as a communication option for your Arduino projects.

315 Mhz transceivers

Short-range RF Transceivers like the 433MHz transmitters discussed in one of our previous post are very popular among DIY hobbyist and makers, however, their short-range was a bottleneck for users as they barely offer signal coverage for an area bigger than a standard room. To solve this and give makers more option and value, PMD Way developed these new 315/415 Mhz transceivers.

The new transceivers work the same way as the cheaper modules, maintaining the same pin configuration/layout, support the same libraries (like the VirtualWire library for Arduino) but the new modules, however, have a greater power output and a solid, convenient PCB antenna which increases range while also keeping things neat. The attainable transmitting power of the modules depends on the voltage applied at it’s VCC pin. For instance, when VCC is 5v, the modules are able to reach 150m range in open space, as such, the range can be improved further by increasing the supply voltage.

Some of the specifications of the long-range RF Modules are highlighted below:

  • Transmitter working voltage: 4~12VDC
  • Transmitter working current: 40mA
  • Transmit power: 27dBm @ 12VDC
  • Working frequency: 315MHz
  • Transfer rate: 4.8kbps (max)
  • Transmission distance: When supply power is 5V, the open area transmission distance is around 150 meters. To increase transmit range, you can increase the power to the transmitted up to 12V DC.
  • Antenna: Onboard 315MHz dedicated PCB Antenna.
  • Receiver operating voltage: 2~5.5V DC
  • Receiver current: 2mA
  • Sensitivity: -110db

For today’s tutorial, we will demonstrate how to use these modules by creating a simple sketch to send messages between two Arduino fitted with these modules. The first Arduino board will serve as the transmitter using a pushbutton such that when the push button is pressed, it sends an on/off data via the long-range RF modules to the second Arduino which serving as the receiver, which will turn the LED connected to it on/off in accordance with the signal received.

Required Components

The following components are required to build this project;

  1. 315MHz Long Range Wireless RF Module (1 Set of Transmitter and Receiver)
  2. Arduino Uno or compatible Boards (2)
  3. Breadboards (2)
  4. Pushbutton (1)
  5. Jumper Wires
  6. 330 Ohms
  7. LED (1)
  8. An external power supply for one Arduino. This can be an AC-DC wall wart, USB power bank, 9V battery, any power source to power the transmitter.

Two Arduino boards are required because we will build a complete system with a transmitter and receiver. If you don’t have an Arduino Uno available feel free to use any Arduino compatible board.

Schematics

As mentioned during the introduction, we will essentially build two projects. One will serve as the transmitter while the other will serve as the receiver as such, we have two schematics. The image of the short-range RF modules will be used since the connection is the same and the Long-range RF modules do not have a fritzing package.

Transmitter Schematics

The transmitter schematics is quite simple, we only need to connect a push-button and the transmitter of the 315MHz transceiver to the Arduino as shown in the image below.

Transmitter Schematics

The pin-to-pin connection between the Arduino and the RF transmitter module is described below:

Arduino – RF Transmitter

5v - VCC
GND - GND
D12 - Data

Receiver Schematics

For the Receiver, we will connect a LED to the Arduino in addition to the Receiver of the 315MHz transceiver as shown in the image below.

Receiver Schematic

The pin-to-pin connection between the Arduino and the RF Receiver module is described below:

Arduino – RF Receiver

5v - VCC
GND - GND
D12 - Data

Go over the connections once more to ensure everything is as it should then proceed to the code section.

Code

The Code for this project is heavily dependent on the popular Virtual Wire Library. The library can be installed via the Arduino library manager or by downloading the library from the attached link and installing it by extracting it into the Arduino Libraries folder. The Virtualwire library comprises functions that allow us to perform actions like set the transmission power of the RF modules, encode the data, etc., generally reducing the amount of code we need to write to interact with the RF Modules.

With the library installed we can now proceed to write the code. Just like the schematics, we will write two sketches for this project; one for the transmitter and the other for the receiver.

Transmitter Sketch

We will start with the code for the transmitter. As mentioned during the introduction the sketch for the transmitter will simply send commands (a for ON, b for OFF) to the receiver when off.

We start as usual by importing the libraries that we will use, which in this case, is just the VirtualWire library.

// Transmitter
#include <VirtualWire.h>

Next, we create variables that will be used to hold different data like the characters to be sent by the transmitter, the buffer length, and the buffer name. We also declare the pin of the Arduino to which the push-button is connected.

uint8_t buf[VW_MAX_MESSAGE_LEN];
uint8_t buflen = VW_MAX_MESSAGE_LEN;
const char *on2 = "a";
const char *off2 = "b";
int button = 8;

With that done, we then proceed to the void setup() function. Here we invert the enable signal of the transmitter, use the vw_setup() function to specify the speed of communication in bps, use the vw_set_tx_pin() to specify the pins of the Arduino to which the transmitter pin is connected and finally, set the pin mode of the pin to which the push button is connected as input.

void setup()
{
  vw_set_ptt_inverted(true); // Required for RF Link modules
  vw_setup(300); // set data speed
  vw_set_tx_pin(12);
  pinMode(Button, INPUT);
}

With that done, we move into the void loop function, which is where the real action happens. The void loop function comprises of two if statements which check the state of the push-button. The vw_send() function is called to transmit the appropriate data based on the state of the push button.

void loop()
{
  if (digitalRead(button)==HIGH)
  {
    vw_send((uint8_t *)on2, strlen(on2)); // send the data out to the world
    vw_wait_tx(); // wait a moment
    delay(200);
  }
  if (digitalRead(button)==LOW)
  {
    vw_send((uint8_t *)off2, strlen(off2));
    vw_wait_tx();
    delay(200);
  }
}

The complete code for the transmitter is available below and also attached under the download section of the tutorial.

// transmitter sketch

#include <VirtualWire.h>
uint8_t buf[VW_MAX_MESSAGE_LEN];
uint8_t buflen = VW_MAX_MESSAGE_LEN;
const char *on2 = "a";
const char *off2 = "b";
int button = 8;

void setup()
{
  vw_set_ptt_inverted(true); // Required for RF Link modules
  vw_setup(300); // set data speed
  vw_set_tx_pin(12);
  pinMode(Button, INPUT);
}

void loop()
{
  if (digitalRead(button)==HIGH)
  {
    vw_send((uint8_t *)on2, strlen(on2)); // send the data out to the world
    vw_wait_tx(); // wait a moment
    delay(200);
  }
  if (digitalRead(button)==LOW)
  {
    vw_send((uint8_t *)off2, strlen(off2));
    vw_wait_tx();
    delay(200);
  }
}

Next is the Receiver Sketch.

Receiver Sketch

The sketch for the receiver is quite straightforward. When data is received from the transmitter, the receiver processes the data and issue commands turns the LED ON or OFF depending on the command that was received.

We start the sketch as usual, by including the library that will be used, which is still the same virtual wire library used for the transmitter sketch. Next, we create variables buf and buflen that will be used to store the data received from the transmitter along with its length and also declare the pin of the Arduino to which the LED is connected.

#include <VirtualWire.h>
uint8_t buf[VW_MAX_MESSAGE_LEN];
uint8_t buflen = VW_MAX_MESSAGE_LEN;
int ledpin = 2;

With that done, we write the void setup() function.

We start the function by inverting the state of the enable pin, set the communication speed to 300bps to match that of the transmitter, set the rx pin as D11, start the receiver -> set it to listen for incoming traffic, and then specify the mode of the pin to which the LED is connected.

void setup()
{
  vw_set_ptt_inverted(true); // Required for RF link modules
  vw_setup(300);
  vw_set_rx_pin(11);
  vw_rx_start();
  pinMode(ledpin, OUTPUT);
}

up next is the void loop() function.

The void loop() function is quite straight forward. We check if data was received into the buffer by calling the vw_get_message function. If data was received, a select case statement is then used to compare the data that was received. If “a” was received, the LED is turned ON and if “b” was received, the LED is turned OFF.

void loop()
{
  if (vw_get_message(buf, &buflen))
  {
    switch(buf[0])
    {
    case 'a':
      digitalWrite(ledpin, HIGH);
      break;
    case 'b':
      digitalWrite(ledpin, LOW);
      break;
    }
  }
}

That’s it! The complete code for the receiver is below and also attached under the download section of the project.

// receiver sketch

#include <VirtualWire.h>
uint8_t buf[VW_MAX_MESSAGE_LEN];
uint8_t buflen = VW_MAX_MESSAGE_LEN;
int ledpin = 2;

void setup()
{
  vw_set_ptt_inverted(true); // Required for RF link modules
  vw_setup(300);
  vw_set_rx_pin(11);
  vw_rx_start();
  pinMode(ledpin, OUTPUT);
}

void loop()
{
  if (vw_get_message(buf, &buflen))
  {
    switch(buf[0])
    {
    case 'a':
      digitalWrite(ledpin, HIGH);
      break;
    case 'b':
      digitalWrite(ledpin, LOW);
      break;
    }
  }
}

Demo

Go over the connections once again to ensure all is correct then connect both devices to the computer and upload the sketch one after the other.

With the sketches uploaded, power both devices and press the push-button on the transmitter. You should see the LED connected to the receiver light up.

Demo

The long-range RF modules provide an alternative way to implement cheap communication over somewhat large distances as such, it is worth considering for your low-budget, local network of devices project.

That’s it for today’s project. Feel free to ask any questions you might have about this via the comment section.

A video of the project in action can be found on youtube.

Resources:

RK3399 compute module and carrier follow 96Boards SOM spec

Geniatech and Linaro announced a “SOM 3399” module that adopts the 96Boards SOM spec and runs Linux on a Rockchip RK3399. There’s also a “CBD96-3399” carrier for the module.

Linaro Ltd, the open source collaborative engineering organization developing software for the Arm® ecosystem, and 96Boards Manufacturing Partner Geniatech today announced the launch of another Rockchip RK3399 Board and Carrier Board.

The module adopts the 96Board Computing SOM specification launched in April 2019. The SOM specification is based on a standard form factor and is compatible across SoCs. This means more choice and flexibility for developers, who can seamlessly plug and play between a range of different SOM solutions. The launch of the RK3399 provides yet another SOM solution, delivering low power consumption and high efficiency, with a CPU frequency up to 1.8 GHZ and excellent image processing ability. The RK3399 SOM can also be combined with the carrier board to form a complete industrial application motherboard, which can be applied in various embedded Internet of Things fields.

CBD96-3399 Rockchip RK3399 Carrier Board

The SOM 3399 includes the following features:

  • 6-Core CPU provides high-speed processing capability
  • High performance Dual-core Arm Cortex-A72 MPCore and Quad-core Arm Cortex-A53MPCore processors. Big cluster with dual-core Cortex-A72 is optimized for high performance and little cluster with quad-core Cortex-A53 is optimized for low power.
  • T860MP4 Quad-core GPUdeliversexcellent image processing ability
  • 3D Graphics Engine:ARM® Mali-T860 MP4 Quad-core GPU Support OpenGL ES1.1/2.0/3.0 and OpenCL 1.2, DirectX11.1. Supports AFBCEmbedded 4 shader cores with shared hierarchical tiler.
  • 4Kx2K up to 60fpsHD Resolution
  • Supports 4Kx2K (VP9H.265/HEVC/H.264/AVC), 1080P @ 60fps multi-format video decoding (MPEG-1,MPEG-2,MPEG-4,VC-1),and video encoder for H.264 up to HP@level 4.1, MVC and VP8.  Multiple video input and output interfaces
  • 2 MIPI-DSI input interfaces with two ISP image processors, and Embed two VOPs. Supports dual-screen simultaneous/dual-screen display.
  • Ultra-high integration – ultra-small size

The core board integrates RK3399, CPU DDR, eMMC, a power management module, and Ethernet PHY chip. It has high integration, which greatly reduces the design difficulty of the application backplane and helps enterprises to quickly develop mass production specific application products.

Specifications listed for the CBD96-3399 carrier with SOM 3399 module include:

  • Processor (via SOM 3399) — Rockchip RK3399 (2x Cortex-A72 cores @ up to 1.8GHz, 4x Cortex-A53 cores at up to 1.4GHz); 28nm fab; Mali-T860 GPU
  • Memory/storage:
    • 2GB (or optional 4GB) LPDDR4 RAM (via SOM 3399)
    • 8GB (or optional 16/32/64GB) eMMC (via SOM 3399)
    • MicroSD slot
  • Networking:
    • Gigabit Ethernet port
    • 802.11b/g/n/ac (2.4GHz/5GHz) with Bluetooth 4.2
    • GPS
  • Other I/O:
    • HDMI port for up to 4K UHD H265/H264/VP9 or 3x 1080p@30fps
    • USB 3.0 host port
    • USB Type-C port
  • Expansion:
    • 96Boards 40-pin low-speed (UART x2, SPI, I2S, I2C x2, GPIO x12, DC)
    • 96Boards 60-pin high-speed (LVDS, 4L MIPI-DSI, 2L+4L MIPI-CSI, USB, I2C x2)
    • 60-pin secondary (4L MIPI-DSI, SSC serial, TSIF)
    • 2x analog (headset, speaker via sound-wire, mics, line-outs)
  • Power — 12V DC jack
  • Dimensions — 126 x 124mm
  • Operating system — Android 7.1; Linux

Boards can be purchased directly from Geniatech’s website https://www.geniatech.com/product/som3399/

No pricing or availability information was provided for the SOM 3399 and CBD96-3399. More information may be found in Linaro’s announcement and Geniatech’s SOM 3399 and CBD96-3399 product pages. More details should eventually appear on Linaro’s 96Boards SOM page.

Introducing the Arduino Uno WiFi

A while back in May 2018 at the Maker Faire Bay Area, Arduino after a long year of legal battles announced the launch of two boards which were the MKR Vidor 4000 and the Arduino Uno WiFi Rev 2 which was designed to come up as a somewhat replacement for the legendary Arduino Uno.

This was at a time when several development boards like the ESP32 and NodeMCU which was WiFi enabled were becoming the darling of the maker community as such it quite made sense for Arduino to launch something similar for the beloved Arduino Uno.

The Arduino Uno WiFi Rev2 is basically an Arduino Uno Rev3 with more kick. Unlike the Arduino Uno, it is based on the ATmega4809 which is a microcontroller featuring the 8-bit AVR®  processor with hardware multiplier – running at up to 20MHz and with up to 48 KB Flash, 6 KB SRAM and 256 bytes of EEPROM in 48-pin package. The Uno WiFi Rev2 was the first AVR device to feature Microchip’s Core Independent Peripherals (CIP) and it is one of the key features that make the Uno WiFi a bigger deal than it appears on the surface.

Asides from running on a different processor, the Uno WiFi Rev2 is beyond the Arduino Uno on so many levels as the board comes with more RAM (6KB) and Flash memory (48KB), three hardware UARTS allowing communication which means at least 3 serial devices can be connected, and 6 integrated high-speed Analog-to-Digital Converter (ADC). To show how high up the Uno WiFi is compared to the Uno, it also comes with a Microchip ATECC608A  cryptographic co-processor which provides hardware-based security and hardware-based key storage which are super useful when connecting your projects to the cloud.

To compete with other WiFi boards dominating the Maker Market, the Chief feature for Uno WiFi is probably the WiFi Support as it comes with the NINA-W102 which is the same ESP32-based U-Blox WiFI module used by the new MKR WiFi 1010 announced during the Last Arduino Day.

It is important to note that the Uno WiFi Rev2 is quite different from the Arduino Uno WiFi which was essentially an Arduino Uno with an ESP8266 WiFi module slapped on it but it has since been retired and is no longer being produced.

The Arduino Uno WiFi Rev 2 is currently available for purchase from the Arduino Store.

The BOXER-8300AI Series: Powering AI@Edge with Intel® Movidius™ Myriad™ X

AAEON, the award winning industry leader in AI@Edge solutions, announces the launch of the BOXER-8300AI Series. Including the BOXER-8310AI, BOXER-8320AI, and the BOXER-8330AI (Coming Q4 2019), this family of AI@Edge embedded box PCs power AI and edge computing thanks to the innovative Intel® Movidius™ Myriad™ X.

At the core of the BOXER-8300AI series is the innovative AI Core X module from AAEON. Each AI Core X module features the Intel® Movidius™ Myriad™ X VPU. The Intel® Movidius™ Myriad™ X provides high performance processing, with speeds up to 105 fps (80 fps typical) and 1 TOPS as a dedicated neural network. The BOXER-8300AI Series features improved thermal design, allowing the Intel® Movidius™ Myriad™ X to operate at higher temperatures without loss of performance.

The BOXER-8310AI offers users entry-level value and performance in a compact fanless system. Small enough to fit almost anywhere, this embedded AI@Edge system is powered by a choice of Intel® Pentium™ N4200 or Intel® Celeron™ N3350 processors with up to 8 GB of RAM. The BOXER-8310AI features one AI Core X module, providing users with processing capability to power AI solutions such as traffic monitoring, people counting, or even smart retail solutions. With its improved thermal design, the BOXER-8310AI can operate easily in temperatures from -20°C to as high as 55°C

OXER-8300AI series

The BOXER-8320AI brings the power of the 7th Generation Intel® Core™ i3 Mobile Processor as well as two AI Core X modules, offering processing speeds up to 210 fps (160 typical). The fanless BOXER-8320AI is built with a din-rail mount design and form factor, allowing it to easily integrate into any industrial space, or control cabinet. The advanced thermal design of the BOXER-8320AI allows it to operate in conditions from -20°C to 60°C. With the power of the Intel® Core™ i3 processor and two AI Core X modules, the BOXER-8320AI can be used in high-performance AI applications such as smart security, facial recognition, and more.

The BOXER-8300AI series offers users several advantages over other embedded AI@Edge systems. All BOXER-8300AI systems are compatible with the Intel® Distribution of OpenVINO toolkit, allowing users to run AI inferences on existing frameworks (such as TensorFlow or Caffe) or to create their own. The BOXER-8300AI supports expandable storage, supporting the full suite of Linux and Ubuntu. This also allows the BOXER-8300AI series to be expanded with support for wireless cards such as WiFi, 4G, or Bluetooth. Memory is also easy to replace and upgrade.

AAEON offers all of our customers support to get their projects off the ground. Our manufacturing services and OEM/ODM support provides end-to-end guidance to see your project from concept to production and beyond. For others, AAEON offers complete Linux with Intel® Distribution of OpenVINO toolkit images which can be downloaded from AAEON to get your BOXER-8300AI system up and running the moment you turn it on.

“The BOXER-8300AI series with AI Core X offers our customers a competitive solution to powering AI and Edge Computing,” said Ken Pan, Product Manager with AAEON’s System Platform Division. “With AAEON’s industry leading support, it’s easier than ever for customers to make the choice for an Intel® based AI@Edge platform.”

More information about the product can be found here.

HealthyPi v4 – Unplugged, monitors vital signs using ESP32

HealthyPi v4 is a wireless, wearable, open source vital signs monitor powered by ESP32.

Reliable health monitoring has traditionally required that we tether ourselves to machines capable of recording our vital signs around the clock. Outside of a clinical setting, however, this is rarely practical. We developed HealthyPi v4 in part to address this challenge. Building upon its predecessor, HealthyPi v4 is a fully open source, standalone vital signs monitor with wireless and wearable capabilities.

HealthyPi v4 measures the following parameters in real-time and with high accuracy:

  • Electrocardiogram (ECG) data, heart rate, and heart-rate variability
  • Respiration based on an impedance pneumograph
  • Pulse oximetry (SPO₂)
  • Body temperature

HealthyPi v4 sets a new standard for mobility, wearability, and wireless connectivity. By enabling continuous, real time monitoring of physiological data, it not only provides valuable insight into health and wellness indicators, it opens the door to areas of research that were not previously accessible.

Specifications

  • Microcontroller and wireless connectivity: ESP32, in WROOM32 module format, with a Dual-core Xtensa 32-bit CPU, 4 MB of on-board flash, Wi-Fi, and support for BLE
    • Wireless interface: Wi-Fi and Access Point (AP) modes, a 2.4 GHz radio with an on-board PCB antenna that is compatible with Bluetooth 4.2 and BLE
    • Firmware programming: Supports Arduino IDE as well as Espressif ESP-IDF
  • Sensors:
    • ECG and respiration front end: Texas Instruments (TI) ADS1292R 24-bit analog front end with signal-to-noise ratio (SNR) of 107 dB
    • Pulse oximetry front end: TI AFE4400 pulse oximetry front end with integrated LED driver and 22-bit ADC
    • Temperature sensor: Maxim MAX30208 digital body temperature sensor for monitoring skin temperature
  • Form factor: Raspberry Pi HAT form factor (65 mm X 56 mm)

The project is live on www.crowdsupply.com and has 34 days to go.

TOP PCB Companies